A Recent Systematic Review: System Identification for Modeling and Control in Autonomous Vehicles

A Recent Systematic Review: System Identification for Modeling and Control in Autonomous Vehicles | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Systematic Review A Recent Systematic Review: System Identification for Modeling and Control in Autonomous Vehicles Mohd Zakimi Zakaria, Mohd Sazli Saad, Azuwir Mohd Nor, Mohamad Ezral Baharudin This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5730506/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This systematic literature review aims to identify recent trends and developments in system identification for the modeling and control of autonomous vehicles. Self-driving cars require robust operational dynamics that require modeling to ensure that the vehicles perform complex tasks and respond to changes in the working environment. In response to this, efforts were made to follow the guidelines of the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) checklist. Following pilot testing and database selection, Scopus and Web of Science searches produced 31 primary studies that met the inclusion criteria. These studies are categorised into three themes: The special topics presented include: (1) Autonomous Vehicles and Navigation Control, consisting of recent developments in path planning, obstacle detection, and mode switching; (2) System Identification and Modeling Techniques, which discusses dynamic model identification, real-time parameter estimation, and observer-based methodology; and (3) Machine Learning and Advanced Control Approaches, which discusses the integration of data-driven models, reinforcement learning, and hybrid control systems on vehicles. The findings indicate that integrating conventional control theories with contemporary advanced machine learning reduces reliability, flexibility, and performance. They also highlight how AV should obtain real-time data and IoT to enhance the performance of the control system under conditions of uncertainty. Considering this, this review finds that system identification remains a fundamental area to make breakthroughs in the development of autonomous vehicles because it offers a link between simulation and real-world results. Therefore, the findings offer a guideline for future research focusing toward making control strategies more intelligent and robust with policies for safer and more efficient auto referent systems in land, airborne, and water vehicle systems. System Identification Autonomous Vehicles Mathematical Modeling Machine Learning Advanced Control PRISMA Figures Figure 1 1. Introduction The advancement of autonomous vehicle technology has generated considerable interest in system identification for modeling and control. The identification of systems focuses on building mathematical models of dynamic systems using measured data, which is vital in the design and application of effective control approaches in self-driving cars. This review paper is intended to offer a state-of-the-art survey of the current approaches and uses of system identification for developing models of autonomous vehicles and controlling their movements. Self-driving cars must predict their paths to avoid danger; to do this, they need accurate and sturdy models. He noted that traditional modeling techniques do not work effectively because of the complexity and interference that characterizes vehicle systems. As such, researchers have now shifted toward data-driven and machine-learning methods to increase model viability and stability. There is a specific approach, the Learning Model Predictive Control (LMPC), which improves the decision-making process after considering the outcomes of previous instances. This method has been successfully applied to autonomous racing, where reduction in lap time is one of the most important goals. The LMPC framework employs system identification to build local safe sets and optimal value functions, and greatly decreases the computational load with comparable or improved performance (Kabzan et al., 2019 ; Rosolia et al., 2016 ; Rosolia & Borrelli, 2019 ). Another strategy is linear parameter varying (LPV), which uses machine learning to create mathematical models that account for the dynamic changes in a system. It has proven useful for modeling lateral vehicle dynamics, which improves the path-following control (Fényes et al., 2021 ). The remainder of this paper is organized as follows: The model-based motion planning and control systems described in this article have been developed and include global routing, behavior, and efficient route tracking for many driving scenarios (S. Xu et al., 2022 ). In addition, system identification techniques have been applied to certain subsystems of a vehicle, such as steering and power-train control. For example, it has been demonstrated how generalized least squares algorithms have been applied to depict the steering system of High-speed Intelligent Vehicles and thus contribute to the synthesis of a suitable controller (B. Li et al., 1999 ). Regarding longitudinal control, pioneer controllers have been developed to handle powertrain nonlinearity and provide smooth and responsive vehicle drives (Aziziaghdam & Alankuş, 2022 ). Furthermore, the integration of image processing and system identification provides a basis for control systems for the production of automatic vehicles for agricultural use. These systems harness the features of visuals to guide and eliminate barriers, illustrating how system identification can function on any terrain (Barrozo & Lazcano, 2019 ). Integral sliding-mode control, together with other model-free adaptive control techniques, has expanded the essential functions of autonomous vehicle systems. These strategies utilize data collected from the Internet to construct models for adaptation in a way that realizes high performance even when system models are unavailable or imprecise (D. Xu et al., 2018 ). Finally, the racing nature of autocarts has encouraged better system architectures and sophisticated simulation methods. These tools help enhance the control system parameters to provide superior performance and safety to the vehicle (Culley et al., 2020 ). In summary, system identification for AV modeling and control is advancing with newly proposed and promising ideas and techniques. Each of the four methodologies is described in this review, along with the future trends in this important area of study. Thus, progress in the field of self-driving cars has boosted the growth of an international treaty called System Identification, which deals with the development of mathematical models based on data to improve the control of methods. However, classical modeling approaches are unable to address the nonlinearity and richness of vehicle dynamics, for which data and machine learning solutions are used. Methods such as Learning Model Predictive Control (LMPC) enhance the performance of the control signal by adjusting past results, and hence have been significant when used in applications that include self-driving car racing. Building on the design of the full-path model, Linear Parameter Varying (LPV) models of the system take path- following control to the next level by incorporating changes in system dynamics. Furthermore, model-based planning systems enhance both global routing and behavior planning for various driving situations, and enable trajectory tracking. SID also has a crucial function as a component of sub-schemes that manage turning and powertrain settings in a car, making it run effectively. It does not stop in traditional transportation because agricultural vehicles also use image processing for location determination. Nonlinear adaptive techniques, such as sliding-mode control, will provide a robust solution in applications where the environment is continuously dynamic. Self-driving racing continues to progress from simulation, parameters, and added value to performance and safety. Regardless of the research advancements, system identification continues to be critical for the development of effective self-driving cars in multiple applications. 2. Literature Review The identification of model systems is particularly important in the modeling and control of autonomous vehicles in order to establish reliable control systems. Several techniques have been employed to overcome obstacles to system identification in this field. The most evident method is the Learning Model Predictive Control (LMPC) method, which has been used in Autonomous Racing. Here, we use the information accrued from the previous laps to refine the controller continually while simultaneously not adding to race time. The system identification strategy in this context involves constructing an affine time- varying prediction model using vehicle kinematic equations showing the efficiency in experimental results on the Berkeley Autonomous Race Car (BARC) platform (Rosolia & Borrelli, 2019 ). Similarly, another study proposed the application of a learning-based model predictive control technique in autonomous racing, where vehicle dynamics were estimated from previous laps, and the control performance was optimized (Rosolia & Borrelli, 2019 ). Another new approach for modeling for control design in autonomous vehicles has also been presented, which revolves around generating a control-oriented model within an LPV framework. This method uses feature selection involving scheduling variables and machine learning to select parameters for the LPV model, which maintains a high fitting accuracy on large datasets. The LPV model derived above has been applied to the path-following control of autonomous vehicles (Fényes et al., 2021 ). For high-speed intelligent vehicles, the identification was used to model the steering system. Generalized least-squares algorithms were adopted to obtain the differential equation model, combining the differential equation model to build a satisfactory control system for the ANIVS system (B. Li et al., 1999 ). Furthermore, a model-free adaptive control scheme was proposed for an autonomous four- wheeled mobile vehicle (4WMV) parking system. This approach includes online identification of the data- driven model, as well as an integrated sliding-mode controller, which has better performance than traditional control methods (D. Xu et al., 2018 ). In addition, a guaranteed under-impact and real-time motion planning and control framework for an autonomous vehicle was developed in a real-world experiment integrating global routing, behavior planning, local trajectory planning, and tracking. More precisely, this system has been verified for different complex traffic conditions, as it is effective and accurate (S. Xu et al., 2022 ). Another study aimed at the development of a longitudinal controller for autonomous vehicles with nonlinear power-train characteristics using the reverse plant model in the low-speed range (Aziziaghdam & Alankuş, 2022 ). Therefore, system identification methods are crucial for designing feasible modeling and control methodologies for automated automobiles. These methods start with learning-based techniques and data-intensive modeling and extend to model-free adaptive control of autonomous vehicles to enhance their accuracy, efficiency, and performance in different scenarios of road operation (Aziziaghdam & Alankuş, 2022 ; Fényes et al., 2021 ; B. Li et al., 1999 ; Rosolia et al., 2016 ; Rosolia & Borrelli, 2019 ). Nevertheless, there are several issues with system identification for AVs that still present challenges. For example, poor weather conditions act as major barriers to AV sensors, causing the system to be poor (Chalvatzaras et al., 2023 ). Second, the need for accurate localization techniques, whereby system identification essentially lies, has provoked different map-based and probabilistic methods (Vargas et al., 2021 ). The nature of AV sensors with regard to adverse climatic conditions and the proliferation of variables in real-life traffic situations requires constant innovation and development. Research into human-like driving systems, which explores ways to minimize the gap between a machine decision and a human decision, also contributes to refining system identification by making AV more sensitive to real conditions (Dasarathy, 1980 ; L. Li et al., 2018 ). Mathematical modeling is instrumental in the design and supervision of self-driving cars and underlies different features of the vehicle, including guiding, routing, and the vehicle itself. These include the application of mathematical modeling to examine various aspects of the enhanced performance and safety of automobiles. For example, Petrov and Nashashibi ( 2014 ) described a mathematical model and an adaptive controller for autonomous vehicle the overtaking, with a focus on the real-time generation of overtaking trajectory and feedback controller in view of the unknown velocities of the overtaken vehicles. Similarly, Namngam et al. ( 2022 ) presented a model for autonomous vehicles to optimize their movement in closed areas, which is useful in agricultural applications where reducing unnecessary path travel reduces cost. As for vehicle dynamic simulations, James et al. ( 2020 ) compared physical and data-driven models under real-world driving and found that physical models are less accurate and more computationally expensive compared to data-driven models, especially neural network models. As discussed in (Fényes et al., 2021 ), this study introduces a new data-driven modeling approach based on the application of machine learning optimization to LPV architectures. These models are useful for the control design of path-following features in self-driving cars. Moreover, Moriwaki and Tanaka ( 2007 ) analyzed the intercessions between vehicle motions and presented a collective mathematical model that is ideal for the entire motion of autonomous passenger cars. Mercy et al. ( 2018 ) proposed an optimization method for motion planning in a dynamic environment to design a smooth motion trajectory without collisions. This is substantiated by the vivid presentation of simulations and experimental evaluation of a mainframe to allow real-time control of this approach. Liniger et al. ( 2015 ) investigated a novel nonlinear model predictive controller (NMPC)-based control strategy in the context of autonomous racing. It enables the car to ride perfectly on the track, dodge barriers, and others in real time. The results not only demonstrate the incredible efficiency of NMPC, but also the practicality of its real-time implementation in high-speed dynamic racing conditions. Lastly, Berrada and Leurent ( 2017 ) outlined the challenges and applications of AV modeling with a comparison of many modeling studies covering spatial and socio-economic impacts. In this paper, we elaborate on the agent-based framework of agent technical description and geography of the system, mathematical instruments in regard to market share, and profitability analysis performance study of the project. Further, Nakamura and Sakakibara (2019) focused on group control strategies for autonomous vehicles that determine the safety of the group control algorithm by the formal verification method and mathematical optimization indicating that group control is vital for future smart cities. Contemporary developments in the mathematical methods of autonomous vehicles include a variety of uses in the determination of vehicle dynamics, control of design, path calculations, and social and economic implications. Taken together, these studies can help advance better, safer, and more dependable autonomous vehicle systems. The ideal means for creating highly accurate and effective vehicular systems that control autonomous vehicles is system identification, which is the mechanized process of identifying, analyzing, and describing vehicular systems. Algorithms such as Learning Model Predictive Control (LMPC) iterate over data to optimize what they do and are a better fit for fast applications such as autonomous racing. Others are the previously mentioned LPV model that improves path-following control by optimizing via learning. For object control in real-time data conditions, adaptive and model-free control methods have demonstrated superiority over traditional methods in parking and steering systems. The mathematical mode expands upon the control design and powertrain, and trajectories that optimize the system when operating in complex scenarios, as well as behavior performance. However, variations owing to environmental factors, such as weather conditions and sensor failures, require additional research. These sophisticated yet more efficient systems for interactive driving of a self-driving car combine model human-like driving, neural networks, and optimization techniques into themselves. These models are also being used for navigation, motion planning, and socio-economic analysis, indicating the growing relevance of system identification toward autonomous driving systems. 3. Materials and Methods The materials and methods assembled in this study were designed to present a strong and reliable framework for this research, implemented over the course of four main phases. The first phase involved the preparation of a comprehensive and methodical search strategy used to interrogate major databases for relevant studies. The next step is screening, which filters out duplicate records and eliminates irrelevant material. The next phase is eligibility, a filter for confirming that each of the included items satisfies the clearly articulated inclusion criteria. The third phase is data extraction and analysis, which involves pulling key data, running synthesis, and interpreting findings. This was in place to maintain the rigor of the research, grounding the results accurately and reliably. 3.1 Identification In this study, important steps of the systematic review process were used to collect a large volume of relevant literature. The approach began with keyword selection, followed by a search for comparable terms in dictionaries, thesauri, encyclopedias, and previous research. All relevant phrases were identified, and search strings were generated for the Web of Science (WoS) and Scopus databases (see Table 1 ). The first phase of the systematic review identified 318 papers relevant to the study issue from the two databases. Table 1 The search string. Scopus TITLE-ABS-KEY ( "system identification" AND model* AND control* AND autonomous AND vehicle ) AND ( LIMIT-TO ( LANGUAGE, "English" ) ) AND ( LIMIT-TO ( PUBSTAGE, "final" ) ) AND ( LIMIT-TO ( DOCTYPE, "ar" ) ) AND ( LIMIT-TO ( PUBYEAR, 2020 ) OR LIMIT- TO ( PUBYEAR, 2021 ) OR LIMIT-TO ( PUBYEAR, 2022 ) OR LIMIT- TO ( PUBYEAR, 2023 ) OR LIMIT-TO ( PUBYEAR, 2024 ) ) Date of Access: October 2024 WoS "system identification" AND model* AND control* AND autonomous AND vehicle (Topic) and 2024 or 2023 or 2022 or 2021 or 2020 or 2019 (Publicati on Years) and Article (Document Types) Date of Access: October 2024 3.2 Screening In the screening phase, each research item was carefully checked to ensure alignment with the defined research questions. This process is zero in studies pertinent to system identification for modeling and control within autonomous vehicles. The duplicate entries were systematically removed. From the initial pool, 235 studies were excluded, narrowing the selection to 83 papers for a deeper evaluation based on strict inclusion and exclusion criteria as shown in Table 2 . The review concentrated on English-language journal articles published between 2019 and 2024, deliberately setting aside conference papers, books, reviews, and any work still pending publication. Finally, 25 duplicates were excluded. Table 2 The selection criterion is searching Criterion Inclusion Exclusion Language English Non-English Timeline 2019–2024 < 2019 Literature type Journal (Article) Conference, Book, Review Publication Stage Final In Press 3.3 Eligibility During the eligibility phase, the third step (58 items) was chosen for an additional examination. At this juncture, the titles and principal content of each article were meticulously assessed to confirm their compliance with the inclusion criteria and alignment with the research objectives. Consequently, 27 papers were removed because they were outside the scope, possessed irrelevant titles, or presented abstracts unrelated to the objectives of the study. This procedure resulted in 31 papers for the final evaluation. 3.4 Data Abstraction and Analysis This study evaluated and synthesized several quantitative research designs using an integrated analysis. The main objective was to identify pertinent themes and subthemes aligned with the study subjects. The first phase involved collecting relevant data. As shown in Fig. 1 , the authors systematically reviewed 31 papers to extract critical information relevant to the research focus on system identification for modeling and control in autonomous vehicles, as listed in Table 3 . To ensure comprehensive coverage, this study assessed the methodologies and findings of multiple investigations. The process of refining themes required collaboration among the authors, as each theme was developed based on data collected within the study environment. A detailed notebook was maintained throughout the process of documenting reflections, interpretations, challenges, and evolving ideas during the data analysis. In the final phase, the authors cross-checked their findings to resolve any inconsistencies, and any disagreements regarding the themes were addressed through team discussions. The selected themes were carefully refined to ensure analytical consistency. This study addressed two key research questions: How can advanced navigation-control strategies improve the operational efficiency and safety of autonomous vehicles in dynamic environments? What role does machine learning play in enhancing the adaptability and robustness of control systems in autonomous vehicle applications? Table 3 The primary data No Authors (Year) Title Year Source title Scopus WoS 1 Jiang et al. ( 2022 ) Path-following control of autonomous ground vehicles based on input convex neural networks 2022 Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering / / 2 Ariza Ramirez et al. ( 2020 ) Exploration of the applicability of probabilistic inference for learning control in underactuated autonomous underwater vehicles 2020 Autonomous Robots / / 3 Wang et al. ( 2022 ) Real-time parameter identification of ship maneuvering response model based on nonlinear Gaussian Filter 2022 Ocean Engineering / 4 Kim et al. ( 2024 ) Field experiment of autonomous ship navigation in canal and surrounding nearshore environments 2024 Journal of Field Robotics / / 5 Wiberg et al. ( 2024 ) Sim-to-real transfer of active suspension control using deep reinforcement learning 2024 Robotics and Autonomous Systems / / 6 Humphreys et al. ( 2020 ) Advancing Fusion with Machine Learning Research Needs Workshop Report 2020 Journal of Fusion Energy / / 7 Idros et al. ( 2024 ) Modelling of Vehicle Longitudinal Dynamics for Speed Control 2024 Journal of Advanced Research in Applied Mechanics / 8 McCrink & Gregory ( 2021 ) Design and Development of a High-Speed UAS for Beyond Visual Line-of-Sight Operations 2021 Journal of Intelligent and Robotic Systems: Theory and Applications / / 9 Rosolia & Borrelli ( 2020 ) Learning How to Autonomously Race a Car: A Predictive Control Approach 2020 IEEE Transactions on Control Systems Technology / / 10 Ahmed et al. ( 2023 ) Survey on traditional and AI based estimation techniques for hydrodynamic coefficients of autonomous underwater vehicle 2023 Ocean Engineering / / No Authors (Year) Title Year Source title Scopus WoS 11 Tischler ( 2022 ) Flight Control Technology Advancements and Challenges for Future Rotorcraft 40th Alexander A. Nikolsky Honorary Lecture 2022 Journal of the American Helicopter Society / / 12 Wu et al. ( 2021 ) System identification and controller design of a novel autonomous underwater vehicle 2021 Machines / / 13 Ozdogan & Leblebicioglu ( 2022 ) Design, Modeling, and Control Allocation of a Heavy-Lift Aerial Vehicle Consisting of Large Fixed Rotors and Small Tiltrotors 2022 IEEE/ASME Transactions on Mechatronics / 14 Kawamura et al. ( 2021 ) Development of small autonomous surface vehicle implementing position control system using sliding mode control 2021 Sensors and Materials / / 15 Lai & Le ( 2021 ) Adaptive Learning-Based Observer with Dynamic Inversion for the Autonomous Flight of an Unmanned Helicopter 2021 IEEE Transactions on Aerospace and Electronic Systems / / 16 Mladenov & Best ( 2023 ) Characterisation of driver longitudinal behaviour using an Unscented Kalman Filter 2023 Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering / / 17 Chen & Ozay ( 2022 ) Data-Driven Computation of Robust Control Invariant Sets With Concurrent Model Selection 2022 IEEE Transactions on Control Systems Technology / / 18 Yan et al. ( 2024 ) Integration of Vehicle Dynamic Model and System Identification Model for Extending the Navigation Service under Sensor Failures 2024 IEEE Transactions on Intelligent Vehicles / / 19 Martinsen et al. ( 2022 ) Reinforcement learning-based NMPC for tracking control of ASVs: Theory and experiments 2022 Control Engineering Practice / / 20 Lee et al. ( 2024 ) Nonlinear Model Predictive Control with Obstacle Avoidance Constraints for Autonomous Navigation in a Canal Environment 2024 IEEE/ASME Transactions on Mechatronics / / 21 Soitinaho & Oksanen ( 2023 ) Local Navigation and Obstacle Avoidance for an Agricultural Tractor With Nonlinear Model Predictive Control 2023 IEEE Transactions on Control Systems Technology / No Authors (Year) Title Year Source title Scopus WoS 22 Ivler et al. ( 2022 ) System identification of rover dynamics: a comparison of three model structures 2022 International Journal of Modelling, Identification and Control / 23 T.K et al. (2020) System identification of flybar-less rotorcraft UAV 2020 Aircraft Engineering and Aerospace Technology / / 24 Souza et al. ( 2023 ) A Convolutional System Identification Approach Mixing Optimal Parameter Estimation and Deep Learning 2023 International Journal of Control, Automation and Systems / / 25 Vicente et al. ( 2021 ) Linear System Identification Versus Physical Modeling of Lateral-Longitudinal Vehicle Dynamics 2021 IEEE Transactions on Control Systems Technology / / 26 Dubey et al. ( 2022 ) Steering model identification and control design of autonomous ship: a complete experimental study 2022 Ships and Offshore Structures / 27 Grip et al. ( 2020 ) Modeling and identification of hover flight dynamics for NASA's Mars helicopter 2020 Journal of Guidance, Control, and Dynamics / / 28 Esmaeili & Modares ( 2024 ) Risk-Informed Model-Free Safe Control of Linear Parameter- Varying Systems 2024 IEEE/CAA Journal of Automatica Sinica / / 29 Deng et al. ( 2021 ) Identification of an Autonomous Underwater Vehicle hydrodynamic model using three Kalman filters 2021 Ocean Engineering / / 30 Gokul Krishnan et al. ( 2024 ) Real-Time Experimental Evaluation and Analysis of PID and MPC Controllers Using HIL Setup for Robust Steering System of Autonomous Vehicles 2024 IEEE Access / / 31 Ma et al. ( 2023 ) Local Learning Enabled Iterative Linear Quadratic Regulator for Constrained Trajectory Planning 2023 IEEE Transactions on Neural Networks and Learning Systems / / 4. Result and Finding The study’s results and findings were organized into three primary themes. The initial theme, autonomous vehicles and navigation control, delves into the latest developments in real-time navigation strategies, path planning, and vehicle autonomy. The second theme, system identification and modeling techniques, emphasizes the importance of parameter estimation and model validation in the development of precise models of dynamic systems. The final theme, machine-learning and advanced control approaches, investigates the integration of data-driven algorithms with control systems to improve performance, adaptability, and decision making. These themes collectively offer a comprehensive perspective on the study's primary areas, emphasizing significant trends and insights into contemporary control engineering. 4.1 Autonomous Vehicles and Navigation Control Various novel approaches and bodies of work for system identification are evident in the literature geared toward modeling and performing control of autonomous vehicles with different levels of autonomy ranging from human-run vehicle applications, such as agricultural tractors, autonomous surface vehicles (ASVs), and underwater vehicles. Considerable attention has been paid to the applications of nonlinear model predictive control (NMPC) and sliding mode control, which are both effective for navigation and obstacle avoidance. Soitinaho and Oksanen ( 2023 ) demonstrated how NMPC could be used in agricultural tractors, for example, when performing path tracking and obstacle avoidance. The experiments indicated that the NMPC technique maintains a cross-track error (xte) under 0.05 meters, demonstrating its physical world operating relevance. Similarly, Lee et al. ( 2024 ) used NMPC to improve the navigation performance of a small cruise boat operating in a canal-structured environment. Using experimental data and system identification, their setup validated NMPC for dealing with nonlinear dynamics and real-time LiDAR-induced obstacle avoidance. These results underscore the need for advanced control strategies to improve navigation systems in autonomous vehicles. An additional promising approach is the establishment of ASVs, developed by Kawamura et al. ( 2021 ). The authors proposed a sliding mode control system with robust properties against model mismatch and external disturbances. The ASV successfully navigated with minimal human supervision, indicating that the system could be trusted in perpetuating aquatic environments. Correspondingly, Kim et al. ( 2024 ) reported a study in which ASV autonomous capabilities were enhanced using additional sensors and control algorithms. The results from the field experiment showed that the system was able to sense its environment and autonomously make complex movements around obstacles, reaffirming sliding mode control as a practical controller in marine applications. Ariza Ramirez et al. ( 2020 ) studied the use of probabilistic inference learning for autonomous underwater vehicles (AUVs) control. Conventional methods of controller tuning that gain some win in stationary backgrounds lose control when conditions are dynamic and practical, as mentioned in their study. Instead, they proposed a probabilistic model-based reinforcement learning procedure that can be adapted to different missions with almost no pre-programming and shows good promise for autonomous navigation of AUVs in uncertain environments. This drastically improves operational efficiency because fewer field trials are needed to find good navigation policies. The other is the ongoing development of flight control technologies for rotorcraft. In his review of the flight control system (Tischler, 2022 ); he illustrated the help of time-domain and frequency-domain analyses to attain stability and handling quality for autonomous flight. Advanced feedback control systems have become essential to consistently meet the performance objectives of many rotorcraft applications such as urban air mobility. These developments illustrate the continued need for system identification and adaptive control to address the problems caused by new classes of vehicles. While modeling such systems can be challenging owing to the complexity of vehicle dynamics and motion, their importance as reliable control systems has sparked increasing interest in the system identification literature around AVs. Therefore, a potential path is derived from neural network design, namely input convex neural networks (ICNNs), which can provide unique benefits for certain path-following control tasks. Traditional modeling methods for AGVs are based on first principles are often impractical, which motivates a shift from model-based approaches to data-driven methods (Jiang et al., 2022 ). The use of the properties of ICNNs transforms the predictive control problem into a convex optimization problem, thus allowing us to compute feasible solutions. An online learning algorithm periodically enhances the model to adapt it to various road conditions and disturbances, showing its effectiveness in simulations. One of the critical components of the system identification problem is the difficulty in identifying the steering dynamics between autonomous ships. An approach of system identification using input-output for a steering model was developed by Dubey et al. ( 2022 ), which was used to identify the models for autonomous vessels. They then adopted models of similar traditional hydrodynamic modeling that lacked proper knowledge, which made it difficult to create efficient control systems. To investigate the behavior of vessels, they derived an autonomous ship model that has self-propelled autonomous capabilities and is equipped with IoT features to allow real-time data acquisition under different conditions. The experimental results highlight the viability of using IoT for hydrodynamic experiments and demonstrate that making data available can help improve control system performance. To enhance the accuracy of system identification techniques for AVs, a combination of state-of-the-art computational methods and real-time data access is required. Martinsen et al. ( 2022 ) and Jiang et al. ( 2022 ) highlight that machine learning to privilege control strategy optimization will require higher degrees of manifestation, which is expected with the technological advancement cycle. Integrating methodologies from classical control theory, such as optimal and model predictive control upper bounds on task performance, with current AI design approaches (reinforcement learning and neural networks) will ultimately lead to autonomous vehicles that learn new skills while still conforming to their original signal bounds. Furthermore, Dubey et al. ( 2022 ) highlighted this gap and called toward more realistic experimental setups that help design better control systems in real-world autonomous maritime applications. A literature review on system identification for autonomous vehicles shows significant contributions associated with computational models in terms of increasing the accuracy of modelled data and nonlinear behavior of control objects. Using machine learning approaches to identify systems has great potential for achieving the best performance and safety in AVs. Nonlinear model predictive control, sliding mode control, and learning-based approaches are state-of-the-art methodologies that aim to improve the navigation and control functionalities of autonomous vehicles. This highlights the need for significant system identification, testing, and control approaches to adapt to unknown or less than perfect scenarios with environmental dynamics. Further research should be conducted to continue developing these methodologies and applying them to other vehicle types and operational scenarios to develop safer and more efficient autonomous systems. With the development of autonomous technologies, such a model can also lead to innovative solutions that can deal with problems arising from newly encountered scenarios. 4.2 System Identification and Modeling Techniques System identification is an important step in the modeling and control of autonomous vehicles, paving the way for the development of predictive models that are essential for control and navigation. The importance of this differential activation and contributions from multiple methodologies in the literature emphasize the complexity of system performance across diverse platforms and environments through effective responses. A rapid yet realistic mathematical model for vehicle longitudinal dynamics is essential for the speed control task and was highlighted by Idros et al. ( 2024 ). The detailed model proposed in this work is integrated with the dynamics of the vehicle body, powertrain, and braking, and simulation was performed using MATLAB Simulink to analyze various control strategies. The results reveal that the hierarchical PID control structure is capable of accurately tracking conventional urban drive cycles, thus validating the model as a platform for further control strategy development. For example, in autonomous navigation, to achieve better model fidelity, Wang et al. ( 2022 ) presented a method based on a nonlinear Gaussian filter for real-time parameter identification that can be utilized to adjust ship maneuvering response models. This investigation illustrates how the addition of observers can reduce parameter drift, leading to enhanced system identification performance. In contrast, these include the analysis of three model structures for rover dynamics and are designed with a higher fidelity model, reaching the conclusion that roll-yaw is the most accurate yet most complex of all others, whereas simpler models, such as dynamic bicycle models, cannot have the totality of other simplified characterizations (Ivler et al., 2022 ). Together, these studies communicate the benefit of an appropriate model structure, given that one has estimated prediction error and must detect a balance between overfitting (by using complex models) and underfitting in dynamic landscapes. In this area, the literature mirrors the advancements in data-driven methods for system identification. Chen and Ozay ( 2022 ) proposed an approach for synthesizing robust control-invariant sets and highlighted the link between model selection and optimal system identification from data. Through their approach, they showed that you can significantly improve the estimates of what makes control systems robust to maintain performance in uncertain environments. Similarly, Vicente et al. ( 2021 ) compared the lateral and longitudinal vehicle dynamics linear system identification with physical modeling. In addition, they concluded that data-driven models can outperform traditional physical models, at least when it comes to following real-world driving behavior, leading them away from more constrained methods of modeling autonomous vehicle systems. Deng et al. ( 2021 ) explored this line of research further. This study employs different Kalman filter algorithms to identify hydrodynamic models for autonomous underwater vehicles and shows that optimized methods, such as the optimized unscented Kalman filter, can help improve the parameter estimation accuracy in noise. This finding is consistent with those of Yan et al. ( 2024 ). Finally, they proposed a sensor-free localization approach combining system identification with vehicle dynamic models. These experiments show that not only is the localization accuracy greatly improved (especially during sensor failure scenarios) but also highlighted the capability of system identification to improve resilience in autonomous systems. Finally, Ma et al. ( 2023 ) used iterative and machine-learning techniques. However, this is the only example of a local learning-enhanced iterative linear quadratic regulator for trajectory planning. This demonstrates the potential of tighter coupling between system identification and model predictive control frameworks to achieve more efficient and flexible trajectory generation approaches for robotic and autonomous systems. Grip et al. ( 2020 ) used a different modeling approach and dealt with challenges due to the Martian environment for the dynamic model of NASA's Mars helicopter and showed that proper system identification techniques are crucial when flying under extreme conditions. Overall, the literature on system identification for autonomous vehicles indicates a shift toward fusing different modeling methodologies to achieve more accurate and robust models while accounting for rich dynamic behavior. This shift toward data-driven approaches represents a larger trend in the literature in which experimental data are used to enhance model fidelity and control performance, making such systems more viable for autonomous navigation. 4.3 Machine-Learning and Advanced Control Approaches The introduction of machine learning and advanced control approaches has led to recent advances in autonomous vehicles. With further emphasis on autonomous systems that need to be reliable and efficient, researchers have also stressed the best design and control methods. For example, Ozdogan and Leblebicioglu ( 2022 ) recently introduced a new heavy-lift aerial vehicle (HLAV), which is designed with a special configuration of propellers that improves its controllability and requires no use of complex mechanisms. Their study emphasized global second-order sliding mode-based nonlinear dynamic modeling, where closed-loop system identification allows for the estimation of parameters and ultimately results in good trajectory tracking performance. Likewise, other researchers such as Gokul Krishnan et al. ( 2024 ) tackled the problem of steering control systems in autonomous vehicles and compared two traditional controllers, PD and PID, with a recent Model Predictive Control (MPC) formulation. Through this study, they emphasized that real-time testing in diverse driving scenarios is vital for assessing the survivability of the designed controller for safe navigation. In addition, advanced deep learning techniques have been integrated into the system identification area for modeling dynamic systems using a convolutional system identification method (Souza et al., 2023 ). This hybrid approach results in the identification of complex behaviors that allow for more realistic scenarios than conventional models, thereby increasing the reliability of autonomous vehicle controls. Lai and Le ( 2021 ) proposed an adaptive learning-based observer with dynamic inversion for flight control of an unmanned helicopter, where a control subject can utilize adaptive learning mechanisms to enhance the performance of control in a non-static environment. Machine-learning techniques have also been employed to model driver behavior. However, for examples closely related to longitudinal driver behavior characterization in real time, Mladenov and Best ( 2023 ) employed an Unscented Kalman Filter (UKF). Our method not only provides estimates of driver response parameters, but can also be used for insurance practice and autonomous driving systems to simulate different driving styles. Exploration of machine learning methods in driver behavior finds this interdisciplinary nature of advancements in autonomous vehicles where exploratory insights from control theory, data analysis, and behavioral modeling are used to predict the risk associated with driving. Finally, we discuss how machine learning can be used in wider areas, such as fusion energy research, as a new tool. Humphreys et al. ( 2020 ) addressed Fusion Energy Challenges With ML/AI Integration: Bridging Theory and Practice with Data-Driven Approaches. These multidisciplinary efforts continue to highlight the strides in tuning control and emphasize the inclusion of machine learning for better accuracy in autonomous systems. This journal focuses on advancing the dynamic interplay between controls, systems, and machines. System identification methods have been widely applied to the control of autonomous vehicles, particularly in machine learning and nonlinear model predictive control. One specific example is the development of risk-informed model-free control strategies (Esmaeili & Modares, 2024 ). They proposed a probabilistic safe control design for linear parameter-varying (LPV) systems that represents safety given closed-loop data and avoids model details. They demonstrated that the variance of a closed-loop system and the probability of safety are both inherently tied to the decision variables and noise covariances. This novel framework of risk-averse control design approaches safety by reducing the probability of safety violations with a proper data- driven control analysis. These safe control systems are necessary to provide solutions that account for the uncertainty in controlling autonomous vehicles. An additional major effort in this area comes from Wiberg et al. ( 2024 ) on heavy vehicles with active suspension systems using deep reinforcement learning (DRL) controllers and studying the sim-to-real transfer of DRL controllers. Unlike traditional robotic work that emphasizes lighter robots, their study faces unique challenges, such as the complicated behavior of a hydraulic driveline and slow actuation. Using system identification and sim-to-real mitigation techniques, such as domain randomization and action delays, the sim-to-real gap can be closed by fine-tuning the simulation parameters. Their findings suggest that training policies can be designed to provide not only simulation performance but also fidelity in real-world motion trajectories. This highlights the need for reliable modeling and simulation methods to obtain reliable control strategies for autonomous systems in difficult environments. Ahmed et al. ( 2023 ) focused on determining the hydrodynamic coefficients of an autonomous underwater vehicle (AUV). Such a thorough survey is relevant and emphasizes that robust dynamic modeling will always be an important prerequisite for designing informative control systems in the case of AUVs moving in six degrees of freedom. This study compares traditional estimation methods (Analytical and Semi-Empirical approaches and CFD) to more recent artificial intelligence techniques (i.e., supervised machine-learning algorithms). The authors discussed the benefits of these newer methods, which are more accurate and computationally cheaper than older generation techniques. In addition, it explains the development of estimation methods in the last 75 years and their relevance in efficiently designing AUV control systems. Researchers have provided important insights into the enhancement of control strategies for autonomous marine vehicles by combining traditional methods with modern artificial intelligence techniques. A strong interaction is identified between model-driven and machine-learning algorithms to ensure system reliability, performance, and adaptability in the ever-changing environment of autonomous cars, as presented in the literature. Such integration is indispensable for making future autonomous systems safe and efficient. From designs of safe control that are informed by risk, strategies around sim-to-real transfer as well as the notions of AI-based estimation methods being proposed collectively in multiple papers reflect a realization of the complex behavior these autonomous vehicles exhibit over their dynamics. Hence, these studies lay the groundwork for innovations to improve the effectiveness and safety of autonomous systems both on land and in water. 5. Discussion and Conclusion Prior work on system identification for autonomous vehicles shows considerable documented development of control strategies for different vehicle types, such as agricultural, marine, and aerial systems. In particular, we have struggled with nonlinear model predictive control (NMPC) and sliding mode control to enhance navigation and obstacle avoidance. In addition, for land and water vehicles, namely NMPC, it is very helpful to tackle the path-tracking problem and environmental disturbances. Sliding mode control ensures the reliability of autonomous surface vehicles (ASVs) owing to their robustness to uncertainties and external disturbances in the environment. Probabilistic learning approaches are adaptable to variations in environments and can generalize their performance from a few field trials, which makes them appropriate for use with autonomous underwater vehicles (AUVs). In state-of-the-art aerial applications, flight control systems have been designed for frequency-domain analysis and adaptive feedback, with a view toward evolving solutions, such as urban air mobility rotorcraft. By using data-driven methods to analyze the results, this study simplifies traditional modeling approaches. Adaptive path-following and handling arbitrary richness scenarios are strongly enabled by Input Convex Neural Networks (ICNNs) with reinforcement-learning algorithms. They offer the advantages of design with models, where IoT-enabling capabilities, help enable data acquisition as required in real time, assisting strong hydrodynamic modeling and efficiency with control systems that are assisted by this. In particular, Advanced Kalman filters for AUV path recovery can also be employed as a method for parameter estimation in noise. Highly specific modeling, such as that for NASA’s Mars helicopter, shows that being able to identify the system precisely is key to performance in very hostile environments. This change to data-centric models is a response to the tendency of machine-learning methods to supersede physical models, increasing adaptability and robustness in systems that operate under dynamic and uncertain conditions. These continuing developments demonstrate the cooperation between machine learning and classic control approaches using resilient adaptive autonomous systems. Closed-loop system identification of a heavy-lift aerial vehicle and nonlinear model advantage for trajectory tracking. Comparative analyses between classical PID controllers and contemporary Model Predictive Control (MPC) highlight the importance of in-situ testing to validate safety and performance during navigation. Driver behavior modeling based on methods using machine learning, such as the Unscented Kalman Filter (UKF), has also found practical use in areas such as insurance assessment and autonomous systems. For this purpose, risk-aware control strategies have recently gained considerable attention to deal with uncertainties and, at the same time, mimicking real-world performance by sim-to-real transfer methods to boost simulation issues in such areas as domain randomization or action delay optimization, contributing to increasing confidence in heavy vehicle control performance focus areas. Future work should focus on developing these methods further and applying them more broadly to a larger class of vehicles and operating conditions as control strategies continue to advance. This will foster the development of secure and trustworthy autonomous systems capable of adapting to complex environments, while driving innovation in control and navigation. The literature indicates a prominent path for merging conventional control theories with data-driven methods and improving the autonomy of a system with higher robustness, flexibility, and efficacy. This evolution has enabled stronger navigation on different platforms and a stable environment in dynamic settings. By integrating conventional control methods with modern AI techniques, this approach provides a robust design framework for enhancing efficiency, safety, and adaptability of autonomous systems, whether land-, air-, or sea-based. Declarations Funding Statement: This study was conducted as part of a personal interest and was not supported by any specific grant. The research aligns with the interests of the Dynamics Modeling & Control System (DyCoS) research group at Universiti Malaysia Perlis (UniMAP). Ethical Compliance This study did not involve any human participants, animals, or other living subjects. It is a recent review focused on system identification methods for modeling autonomous vehicles. Conflict of Interest declaration : The authors declare that they have no affiliations with or involvement in any organization or entity with any financial interest in the subject matter or materials discussed in this manuscript. Author Contribution M.Z. Zakaria wrote the main manuscript text.M.S. Saad prepared Table 1-2.M.E. Baharudin prepared Figure 1.A. Mohd Nor prepared Table 3. References Ahmed, F., Xiang, X., Jiang, C., Xiang, G., & Yang, S. (2023). Survey on traditional and AI based estimation techniques for hydrodynamic coefficients of autonomous underwater vehicle. Ocean Engineering , 268 . https://doi.org/10.1016/j.oceaneng.2022.113300 Ariza Ramirez, W., Leong, Z. Q., Nguyen, H. D., & Jayasinghe, S. G. (2020). Exploration of the applicability of probabilistic inference for learning control in underactuated autonomous underwater vehicles. Autonomous Robots , 44 (6), 1121–1134. https://doi.org/10.1007/s10514-020-09922-z Aziziaghdam, E. T., & Alankuş, O. (2022). Longitudinal Control of Autonomous Vehicles Consisting Power-Train With Non-Linear Characteristics. IEEE Transactions on Intelligent Vehicles , 7 , 133–142. https://doi.org/10.1109/tiv.2021.3115265 Barrozo, J., & Lazcano, V. (2019). Simulation of an Autonomous Vehicle Control System Based on Image Processing. 2019 5th International Conference on Frontiers of Signal Processing (ICFSP) , 88–94. https://doi.org/10.1109/icfsp48124.2019.8938094 Berrada, J., & Leurent, F. (2017). Modeling Transportation Systems involving Autonomous Vehicles: A State of the Art. Transportation Research Procedia , 27 , 215–221. https://doi.org/10.1016/J.TRPRO.2017.12.077 Chalvatzaras, A., Pratikakis, I., & Amanatiadis, A. (2023). A Survey on Map-Based Localization Techniques for Autonomous Vehicles. IEEE Transactions on Intelligent Vehicles , 8 , 1574–1596. https://doi.org/10.1109/TIV.2022.3192102 Chen, Y., & Ozay, N. (2022). Data-Driven Computation of Robust Control Invariant Sets With Concurrent Model Selection. IEEE Transactions on Control Systems Technology , 30 (2), 495–506. https://doi.org/10.1109/TCST.2021.3069759 Culley, J., Garlick, S., Esteller, E. G., Georgiev, P., Fursa, I., Sluis, I. Vander, Ball, P., & Bradley, A. (2020). System Design for a Driverless Autonomous Racing Vehicle. 2020 12th International Symposium on Communication Systems, Networks and Digital Signal Processing (CSNDSP) , 1–6. https://doi.org/10.1109/CSNDSP49049.2020.9249626 Dasarathy, B. (1980). Information Processing For Target Recognition From Autonomous Vehicles . 0219 . https://doi.org/10.1117/12.958535 Deng, F., Levi, C., Yin, H., & Duan, M. (2021). Identification of an Autonomous Underwater Vehicle hydrodynamic model using three Kalman filters. Ocean Engineering , 229 . https://doi.org/10.1016/j.oceaneng.2021.108962 Dubey, A. C., Subramanian, A. V, & Jagadeesh Kumar, V. (2022). Steering model identification and control design of autonomous ship: a complete experimental study. Ships and Offshore Structures , 17 (5), 992–1004. https://doi.org/10.1080/17445302.2021.1889193 Esmaeili, B., & Modares, H. (2024). Risk-Informed Model-Free Safe Control of Linear Parameter-Varying Systems. IEEE/CAA Journal of Automatica Sinica , 11 (9), 1918–1932. https://doi.org/10.1109/JAS.2024.124479 Fényes, D., Németh, B., & Gáspár, P. (2021). A Novel Data-Driven Modeling and Control Design Method for Autonomous Vehicles. Energies . https://doi.org/10.3390/EN14020517 Gokul Krishnan, S., Suresh Kumar, P., Nassar Matara, N., & Wang, Y. (2024). Real-Time Experimental Evaluation and Analysis of PID and MPC Controllers Using HIL Setup for Robust Steering System of Autonomous Vehicles. IEEE Access , 12 , 74711–74723. https://doi.org/10.1109/ACCESS.2024.3406219 Grip, H. F., Johnson, W., Malpica, C., Scharf, D. P., Mandić, M., Young, L., Allan, B., Mettler, B., Martin, M. S., & Lam, J. (2020). Modeling and identification of hover flight dynamics for NASA’s Mars helicopter. Journal of Guidance, Control, and Dynamics , 43 (2), 179–194. https://doi.org/10.2514/1.G004228 Humphreys, D., Kupresanin, A., Boyer, M. D., Canik, J., Chang, C. S., Cyr, E. C., Granetz, R., Hittinger, J., Kolemen, E., Lawrence, E., Pascucci, V., Patra, A., & Schissel, D. (2020). Advancing Fusion with Machine Learning Research Needs Workshop Report. Journal of Fusion Energy , 39 (4), 123–155. https://doi.org/10.1007/s10894-020-00258-1 Idros, S., Abdullah, M., Toha, S. F., & Said, M. S. M. (2024). Modelling of Vehicle Longitudinal Dynamics for Speed Control. Journal of Advanced Research in Applied Mechanics , 123 (1), 56–74. https://doi.org/10.37934/ARAM.123.1.5674 Ivler, C. M., Gowans, N., & Marfise, C. (2022). System identification of rover dynamics: a comparison of three model structures. International Journal of Modelling, Identification and Control , 40 (1), 70–83. https://doi.org/10.1504/IJMIC.2022.124077 James, S. S., Anderson, S., & Lio, M. (2020). Longitudinal Vehicle Dynamics: A Comparison of Physical and Data- Driven Models Under Large-Scale Real-World Driving Conditions. IEEE Access , 8 , 73714–73729. https://doi.org/10.1109/ACCESS.2020.2988592 Jiang, K., Hu, C., & Yan, F. (2022). Path-following control of autonomous ground vehicles based on input convex neural networks. Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering , 236 (13), 2806–2816. https://doi.org/10.1177/09544070221114690 Kabzan, J., Hewing, L., Liniger, A., & Zeilinger, M. (2019). Learning-Based Model Predictive Control for Autonomous Racing. IEEE Robotics and Automation Letters , 4 , 3363–3370. https://doi.org/10.1109/LRA.2019.2926677 Kawamura, Y., Tahara, J., Kato, T., Fujii, S., Baba, S., & Koike, M. (2021). Development of small autonomous surface vehicle implementing position control system using sliding mode control. Sensors and Materials , 33 (3), 883–895. https://doi.org/10.18494/SAM.2021.3216 Kim, J., Lee, C., Chung, D., Cho, Y., Kim, J., Jang, W., & Park, S. (2024). Field experiment of autonomous ship navigation in canal and surrounding nearshore environments. Journal of Field Robotics , 41 (2), 470–489. https://doi.org/10.1002/rob.22262 Lai, Y.-C., & Le, T.-Q. (2021). Adaptive Learning-Based Observer with Dynamic Inversion for the Autonomous Flight of an Unmanned Helicopter. IEEE Transactions on Aerospace and Electronic Systems , 57 (3), 1803– 1814. https://doi.org/10.1109/TAES.2021.3050653 Lee, C., Chung, D., Kim, J., & Kim, J. (2024). Nonlinear Model Predictive Control with Obstacle Avoidance Constraints for Autonomous Navigation in a Canal Environment. IEEE/ASME Transactions on Mechatronics , 29 (3), 1985–1996. https://doi.org/10.1109/TMECH.2023.3312672 Li, B., Wang, R., Zhang, Y., & Wang, Z. (1999). Modeling of steering system of high speed intelligent vehicle by system identification. Proceedings of the IEEE International Vehicle Electronics Conference (IVEC’99) (Cat. No.99EX257) , 243–246. https://doi.org/10.1109/IVEC.1999.830675 Li, L., Ota, K., & Dong, M. (2018). Humanlike Driving: Empirical Decision-Making System for Autonomous Vehicles. IEEE Transactions on Vehicular Technology , 67 , 6814–6823. https://doi.org/10.1109/TVT.2018.2822762 Liniger, A., Domahidi, A., & Morari, M. (2015). Optimization‐based autonomous racing of 1:43 scale RC cars. Optimal Control Applications and Methods , 36 , 628–647. https://doi.org/10.1002/oca.2123 Ma, J., Cheng, Z., Zhang, X., Lin, Z., Lewis, F. L., & Lee, T. H. (2023). Local Learning Enabled Iterative Linear Quadratic Regulator for Constrained Trajectory Planning. IEEE Transactions on Neural Networks and Learning Systems , 34 (9), 5354–5365. https://doi.org/10.1109/TNNLS.2022.3165846 Martinsen, A. B., Lekkas, A. M., & Gros, S. (2022). Reinforcement learning-based NMPC for tracking control of ASVs: Theory and experiments. Control Engineering Practice , 120 . https://doi.org/10.1016/j.conengprac.2021.105024 McCrink, M. H., & Gregory, J. W. (2021). Design and Development of a High-Speed UAS for beyond Visual Line- of-Sight Operations. Journal of Intelligent and Robotic Systems: Theory and Applications , 101 (2). https://doi.org/10.1007/s10846-020-01300-2 Mercy, T., Parys, R. Van, & Pipeleers, G. (2018). Spline-Based Motion Planning for Autonomous Guided Vehicles in a Dynamic Environment. IEEE Transactions on Control Systems Technology , 26 , 2182–2189. https://doi.org/10.1109/TCST.2017.2739706 Mladenov, E., & Best, M. C. (2023). Characterisation of driver longitudinal behaviour using an Unscented Kalman Filter. Proceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering , 237 (14), 3505–3518. https://doi.org/10.1177/09544070221137402 Moher D, Liberati A, Tetzlaff J, A. D. (2009). PRISMA 2009 Flow Diagram. In The PRISMA statement (Vol. 6, p. Moriwaki, K., & Tanaka, K. (2007). Modelling and navigation control of autonomous vehicles. SICE Annual Conference 2007 , 427–434. https://doi.org/10.1109/SICE.2007.4421022 Nakamura, M., & Sakakibara, K. (2019). Formal Verification and Mathematical Optimization for Autonomous Vehicle Group Controllers. 2019 ACM/IEEE 22nd International Conference on Model Driven Engineering Languages and Systems Companion (MODELS-C) , 732–733. https://doi.org/10.1109/MODELS-C.2019.00111 Namngam, K., Chitsakul, P., & Satthamsakul, S. (2022). Mathematical Model for an Autonomous Vehicle to Fulfill the Closed Area. 2022 6th International Conference on Imaging, Signal Processing and Communications (ICISPC) , 133–138. https://doi.org/10.1109/ICISPC57208.2022.00032 Ozdogan, G., & Leblebicioglu, K. (2022). Design, Modeling, and Control Allocation of a Heavy-Lift Aerial Vehicle Consisting of Large Fixed Rotors and Small Tiltrotors. IEEE/ASME Transactions on Mechatronics , 27 (5), 4011–4021. https://doi.org/10.1109/TMECH.2022.3150713 Petrov, P., & Nashashibi, F. (2014). Modeling and Nonlinear Adaptive Control for Autonomous Vehicle Overtaking. IEEE Transactions on Intelligent Transportation Systems , 15 , 1643–1656. https://doi.org/10.1109/TITS.2014.2303995 Rosolia, U., & Borrelli, F. (2019). Learning How to Autonomously Race a Car: A Predictive Control Approach. IEEE Transactions on Control Systems Technology , 28 , 2713–2719. https://doi.org/10.1109/TCST.2019.2948135 Rosolia, U., & Borrelli, F. (2020). Learning How to Autonomously Race a Car: A Predictive Control Approach. IEEE Transactions on Control Systems Technology , 28 (6), 2713–2719. https://doi.org/10.1109/TCST.2019.2948135 Rosolia, U., Carvalho, A., & Borrelli, F. (2016). Autonomous racing using learning Model Predictive Control. 2017 American Control Conference (ACC) , 5115–5120. https://doi.org/10.23919/ACC.2017.7963748 Soitinaho, R., & Oksanen, T. (2023). Local Navigation and Obstacle Avoidance for an Agricultural Tractor With Nonlinear Model Predictive Control. IEEE Transactions on Control Systems Technology , 31 (5), 2043–2054. https://doi.org/10.1109/TCST.2023.3291533 Souza, M. B. A., dos Santos Neto, A. F., de Mello Honório, L., de Oliveira, E. J., Silva, M. F., & Pancoti, A. (2023). A Convolutional System Identification Approach Mixing Optimal Parameter Estimation and Deep Learning. International Journal of Control, Automation and Systems , 21 (8), 2674–2684. https://doi.org/10.1007/s12555- 022-0070-3 Tischler, M. B. (2022). Flight Control Technology Advancements and Challenges for Future Rotorcraft 40th Alexander A. Nikolsky Honorary Lecture. Journal of the American Helicopter Society , 67 (4). https://doi.org/10.4050/JAHS.67.041001 T.K, K. N., V.P, L., & Singh, J. (2020). System identification of flybar-less rotorcraft UAV. Aircraft Engineering and Aerospace Technology , 92 (10), 1483–1493. https://doi.org/10.1108/AEAT-05-2019-0100 Vargas, J., Alsweiss, S., Toker, O., Razdan, R., & Santos, J. (2021). An Overview of Autonomous Vehicles Sensors and Their Vulnerability to Weather Conditions. Sensors (Basel, Switzerland) , 21 . https://doi.org/10.3390/s21165397 Vicente, B. A. H., James, S. S., & Anderson, S. R. (2021). Linear System Identification Versus Physical Modeling of Lateral-Longitudinal Vehicle Dynamics. IEEE Transactions on Control Systems Technology , 29 (3), 1380– 1387. https://doi.org/10.1109/TCST.2020.2994120 Wang, S., Wang, L., Im, N., Zhang, W., & Li, X. (2022). Real-time parameter identification of ship maneuvering response model based on nonlinear Gaussian Filter. Ocean Engineering , 247 . https://doi.org/10.1016/j.oceaneng.2021.110471 Wiberg, V., Wallin, E., Fälldin, A., Semberg, T., Rossander, M., Wadbro, E., & Servin, M. (2024). Sim-to-real transfer of active suspension control using deep reinforcement learning. Robotics and Autonomous Systems , 179 . https://doi.org/10.1016/j.robot.2024.104731 Wu, B., Han, X., & Hui, N. (2021). System identification and controller design of a novel autonomous underwater vehicle. Machines , 9 (6). https://doi.org/10.3390/machines9060109 Xu, D., Shi, Y., & Ji, Z. (2018). Model-Free Adaptive Discrete-Time Integral Sliding-Mode-Constrained-Control for Autonomous 4WMV Parking Systems. IEEE Transactions on Industrial Electronics , 65 , 834–843. https://doi.org/10.1109/TIE.2017.2739680 Xu, S., Zidek, R. A. E., Cao, Z., Lu, P., Wang, X., Li, B., & Peng, H. (2022). System and Experiments of Model- Driven Motion Planning and Control for Autonomous Vehicles. IEEE Transactions on Systems, Man, and Cybernetics: Systems , 52 , 5975–5988. https://doi.org/10.1109/tsmc.2021.3131141 Yan, P., Wen, W., & Hsu, L.-T. (2024). Integration of Vehicle Dynamic Model and System Identification Model for Extending the Navigation Service under Sensor Failures. IEEE Transactions on Intelligent Vehicles , 9 (1), 2236–2248. https://doi.org/10.1109/TIV.2023.3273185 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5730506","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Systematic Review","associatedPublications":[],"authors":[{"id":395919725,"identity":"26d9f812-05dd-4835-b494-c6069ab103c4","order_by":0,"name":"Mohd Zakimi Zakaria","email":"data:image/png;base64,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","orcid":"","institution":"Universiti Malaysia Perlis (UniMAP)","correspondingAuthor":true,"prefix":"","firstName":"Mohd","middleName":"Zakimi","lastName":"Zakaria","suffix":""},{"id":395919726,"identity":"58ad1f58-11d9-49bb-b7c2-e3a35417c226","order_by":1,"name":"Mohd Sazli Saad","email":"","orcid":"","institution":"Universiti Malaysia Perlis (UniMAP)","correspondingAuthor":false,"prefix":"","firstName":"Mohd","middleName":"Sazli","lastName":"Saad","suffix":""},{"id":395919727,"identity":"196ac990-7e71-4d82-82ed-6acbca4256eb","order_by":2,"name":"Azuwir Mohd Nor","email":"","orcid":"","institution":"Universiti Malaysia Perlis (UniMAP)","correspondingAuthor":false,"prefix":"","firstName":"Azuwir","middleName":"Mohd","lastName":"Nor","suffix":""},{"id":395919728,"identity":"cce8d45b-fce1-4647-b436-bdd7ea84e288","order_by":3,"name":"Mohamad Ezral Baharudin","email":"","orcid":"","institution":"Universiti Malaysia Perlis (UniMAP)","correspondingAuthor":false,"prefix":"","firstName":"Mohamad","middleName":"Ezral","lastName":"Baharudin","suffix":""}],"badges":[],"createdAt":"2024-12-29 14:53:04","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5730506/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5730506/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":72674889,"identity":"98c1b32a-4118-4e0b-93c4-d43ef433a2ac","added_by":"auto","created_at":"2024-12-31 06:03:22","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":38499,"visible":true,"origin":"","legend":"\u003cp\u003eFlow diagram of the proposed searching study (Moher D, Liberati A, Tetzlaff J, 2009)\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-5730506/v1/e16a0deff195892b5d2d764c.png"},{"id":72677048,"identity":"6ef917be-8807-403d-a2d2-c3e68c8b2477","added_by":"auto","created_at":"2024-12-31 06:27:24","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":821172,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5730506/v1/0ea9a28a-742c-4169-9750-7945e59d0053.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"A Recent Systematic Review: System Identification for Modeling and Control in Autonomous Vehicles","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eThe advancement of autonomous vehicle technology has generated considerable interest in system identification for modeling and control. The identification of systems focuses on building mathematical models of dynamic systems using measured data, which is vital in the design and application of effective control approaches in self-driving cars. This review paper is intended to offer a state-of-the-art survey of the current approaches and uses of system identification for developing models of autonomous vehicles and controlling their movements. Self-driving cars must predict their paths to avoid danger; to do this, they need accurate and sturdy models. He noted that traditional modeling techniques do not work effectively because of the complexity and interference that characterizes vehicle systems. As such, researchers have now shifted toward data-driven and machine-learning methods to increase model viability and stability. There is a specific approach, the Learning Model Predictive Control (LMPC), which improves the decision-making process after considering the outcomes of previous instances. This method has been successfully applied to autonomous racing, where reduction in lap time is one of the most important goals. The LMPC framework employs system identification to build local safe sets and optimal value functions, and greatly decreases the computational load with comparable or improved performance (Kabzan et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Rosolia et al., \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Rosolia \u0026amp; Borrelli, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAnother strategy is linear parameter varying (LPV), which uses machine learning to create mathematical models that account for the dynamic changes in a system. It has proven useful for modeling lateral vehicle dynamics, which improves the path-following control (F\u0026eacute;nyes et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The remainder of this paper is organized as follows: The model-based motion planning and control systems described in this article have been developed and include global routing, behavior, and efficient route tracking for many driving scenarios (S. Xu et al., \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In addition, system identification techniques have been applied to certain subsystems of a vehicle, such as steering and power-train control. For example, it has been demonstrated how generalized least squares algorithms have been applied to depict the steering system of High-speed Intelligent Vehicles and thus contribute to the synthesis of a suitable controller (B. Li et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e1999\u003c/span\u003e). Regarding longitudinal control, pioneer controllers have been developed to handle powertrain nonlinearity and provide smooth and responsive vehicle drives (Aziziaghdam \u0026amp; Alankuş, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFurthermore, the integration of image processing and system identification provides a basis for control systems for the production of automatic vehicles for agricultural use. These systems harness the features of visuals to guide and eliminate barriers, illustrating how system identification can function on any terrain (Barrozo \u0026amp; Lazcano, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Integral sliding-mode control, together with other model-free adaptive control techniques, has expanded the essential functions of autonomous vehicle systems. These strategies utilize data collected from the Internet to construct models for adaptation in a way that realizes high performance even when system models are unavailable or imprecise (D. Xu et al., \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Finally, the racing nature of autocarts has encouraged better system architectures and sophisticated simulation methods. These tools help enhance the control system parameters to provide superior performance and safety to the vehicle (Culley et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). In summary, system identification for AV modeling and control is advancing with newly proposed and promising ideas and techniques. Each of the four methodologies is described in this review, along with the future trends in this important area of study.\u003c/p\u003e \u003cp\u003eThus, progress in the field of self-driving cars has boosted the growth of an international treaty called System Identification, which deals with the development of mathematical models based on data to improve the control of methods. However, classical modeling approaches are unable to address the nonlinearity and richness of vehicle dynamics, for which data and machine learning solutions are used. Methods such as Learning Model Predictive Control (LMPC) enhance the performance of the control signal by adjusting past results, and hence have been significant when used in applications that include self-driving car racing. Building on the design of the full-path model, Linear Parameter Varying (LPV) models of the system take path- following control to the next level by incorporating changes in system dynamics. Furthermore, model-based planning systems enhance both global routing and behavior planning for various driving situations, and enable trajectory tracking. SID also has a crucial function as a component of sub-schemes that manage turning and powertrain settings in a car, making it run effectively. It does not stop in traditional transportation because agricultural vehicles also use image processing for location determination. Nonlinear adaptive techniques, such as sliding-mode control, will provide a robust solution in applications where the environment is continuously dynamic. Self-driving racing continues to progress from simulation, parameters, and added value to performance and safety. Regardless of the research advancements, system identification continues to be critical for the development of effective self-driving cars in multiple applications.\u003c/p\u003e"},{"header":"2. Literature Review","content":"\u003cp\u003eThe identification of model systems is particularly important in the modeling and control of autonomous vehicles in order to establish reliable control systems. Several techniques have been employed to overcome obstacles to system identification in this field. The most evident method is the Learning Model Predictive Control (LMPC) method, which has been used in Autonomous Racing. Here, we use the information accrued from the previous laps to refine the controller continually while simultaneously not adding to race time. The system identification strategy in this context involves constructing an affine time- varying prediction model using vehicle kinematic equations showing the efficiency in experimental results on the Berkeley Autonomous Race Car (BARC) platform (Rosolia \u0026amp; Borrelli, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Similarly, another study proposed the application of a learning-based model predictive control technique in autonomous racing, where vehicle dynamics were estimated from previous laps, and the control performance was optimized (Rosolia \u0026amp; Borrelli, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAnother new approach for modeling for control design in autonomous vehicles has also been presented, which revolves around generating a control-oriented model within an LPV framework. This method uses feature selection involving scheduling variables and machine learning to select parameters for the LPV model, which maintains a high fitting accuracy on large datasets. The LPV model derived above has been applied to the path-following control of autonomous vehicles (F\u0026eacute;nyes et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). For high-speed intelligent vehicles, the identification was used to model the steering system. Generalized least-squares algorithms were adopted to obtain the differential equation model, combining the differential equation model to build a satisfactory control system for the ANIVS system (B. Li et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e1999\u003c/span\u003e). Furthermore, a model-free adaptive control scheme was proposed for an autonomous four- wheeled mobile vehicle (4WMV) parking system. This approach includes online identification of the data- driven model, as well as an integrated sliding-mode controller, which has better performance than traditional control methods (D. Xu et al., \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2018\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn addition, a guaranteed under-impact and real-time motion planning and control framework for an autonomous vehicle was developed in a real-world experiment integrating global routing, behavior planning, local trajectory planning, and tracking. More precisely, this system has been verified for different complex traffic conditions, as it is effective and accurate (S. Xu et al., \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Another study aimed at the development of a longitudinal controller for autonomous vehicles with nonlinear power-train characteristics using the reverse plant model in the low-speed range (Aziziaghdam \u0026amp; Alankuş, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Therefore, system identification methods are crucial for designing feasible modeling and control methodologies for automated automobiles. These methods start with learning-based techniques and data-intensive modeling and extend to model-free adaptive control of autonomous vehicles to enhance their accuracy, efficiency, and performance in different scenarios of road operation (Aziziaghdam \u0026amp; Alankuş, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; F\u0026eacute;nyes et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; B. Li et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e1999\u003c/span\u003e; Rosolia et al., \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Rosolia \u0026amp; Borrelli, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eNevertheless, there are several issues with system identification for AVs that still present challenges. For example, poor weather conditions act as major barriers to AV sensors, causing the system to be poor (Chalvatzaras et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Second, the need for accurate localization techniques, whereby system identification essentially lies, has provoked different map-based and probabilistic methods (Vargas et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The nature of AV sensors with regard to adverse climatic conditions and the proliferation of variables in real-life traffic situations requires constant innovation and development. Research into human-like driving systems, which explores ways to minimize the gap between a machine decision and a human decision, also contributes to refining system identification by making AV more sensitive to real conditions (Dasarathy, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e1980\u003c/span\u003e; L. Li et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Mathematical modeling is instrumental in the design and supervision of self-driving cars and underlies different features of the vehicle, including guiding, routing, and the vehicle itself. These include the application of mathematical modeling to examine various aspects of the enhanced performance and safety of automobiles. For example, Petrov and Nashashibi (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) described a mathematical model and an adaptive controller for autonomous vehicle the overtaking, with a focus on the real-time generation of overtaking trajectory and feedback controller in view of the unknown velocities of the overtaken vehicles. Similarly, Namngam et al. (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) presented a model for autonomous vehicles to optimize their movement in closed areas, which is useful in agricultural applications where reducing unnecessary path travel reduces cost.\u003c/p\u003e \u003cp\u003eAs for vehicle dynamic simulations, James et al. (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) compared physical and data-driven models under real-world driving and found that physical models are less accurate and more computationally expensive compared to data-driven models, especially neural network models. As discussed in (F\u0026eacute;nyes et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), this study introduces a new data-driven modeling approach based on the application of machine learning optimization to LPV architectures. These models are useful for the control design of path-following features in self-driving cars. Moreover, Moriwaki and Tanaka (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2007\u003c/span\u003e) analyzed the intercessions between vehicle motions and presented a collective mathematical model that is ideal for the entire motion of autonomous passenger cars. Mercy et al. (\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) proposed an optimization method for motion planning in a dynamic environment to design a smooth motion trajectory without collisions. This is substantiated by the vivid presentation of simulations and experimental evaluation of a mainframe to allow real-time control of this approach. Liniger et al. (\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) investigated a novel nonlinear model predictive controller (NMPC)-based control strategy in the context of autonomous racing. It enables the car to ride perfectly on the track, dodge barriers, and others in real time. The results not only demonstrate the incredible efficiency of NMPC, but also the practicality of its real-time implementation in high-speed dynamic racing conditions.\u003c/p\u003e \u003cp\u003eLastly, Berrada and Leurent (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) outlined the challenges and applications of AV modeling with a comparison of many modeling studies covering spatial and socio-economic impacts. In this paper, we elaborate on the agent-based framework of agent technical description and geography of the system, mathematical instruments in regard to market share, and profitability analysis performance study of the project. Further, Nakamura and Sakakibara (2019) focused on group control strategies for autonomous vehicles that determine the safety of the group control algorithm by the formal verification method and mathematical optimization indicating that group control is vital for future smart cities. Contemporary developments in the mathematical methods of autonomous vehicles include a variety of uses in the determination of vehicle dynamics, control of design, path calculations, and social and economic implications. Taken together, these studies can help advance better, safer, and more dependable autonomous vehicle systems.\u003c/p\u003e \u003cp\u003eThe ideal means for creating highly accurate and effective vehicular systems that control autonomous vehicles is system identification, which is the mechanized process of identifying, analyzing, and describing vehicular systems. Algorithms such as Learning Model Predictive Control (LMPC) iterate over data to optimize what they do and are a better fit for fast applications such as autonomous racing. Others are the previously mentioned LPV model that improves path-following control by optimizing via learning. For object control in real-time data conditions, adaptive and model-free control methods have demonstrated superiority over traditional methods in parking and steering systems. The mathematical mode expands upon the control design and powertrain, and trajectories that optimize the system when operating in complex scenarios, as well as behavior performance. However, variations owing to environmental factors, such as weather conditions and sensor failures, require additional research. These sophisticated yet more efficient systems for interactive driving of a self-driving car combine model human-like driving, neural networks, and optimization techniques into themselves. These models are also being used for navigation, motion planning, and socio-economic analysis, indicating the growing relevance of system identification toward autonomous driving systems.\u003c/p\u003e"},{"header":"3. Materials and Methods","content":"\u003cp\u003eThe materials and methods assembled in this study were designed to present a strong and reliable framework for this research, implemented over the course of four main phases. The first phase involved the preparation of a comprehensive and methodical search strategy used to interrogate major databases for relevant studies. The next step is screening, which filters out duplicate records and eliminates irrelevant material. The next phase is eligibility, a filter for confirming that each of the included items satisfies the clearly articulated inclusion criteria. The third phase is data extraction and analysis, which involves pulling key data, running synthesis, and interpreting findings. This was in place to maintain the rigor of the research, grounding the results accurately and reliably.\u003c/p\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Identification\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eIn this study, important steps of the systematic review process were used to collect a large volume of relevant literature. The approach began with keyword selection, followed by a search for comparable terms in dictionaries, thesauri, encyclopedias, and previous research. All relevant phrases were identified, and search strings were generated for the Web of Science (WoS) and Scopus databases (see Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The first phase of the systematic review identified 318 papers relevant to the study issue from the two databases.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe search string.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eScopus\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTITLE-ABS-KEY ( \"system identification\" AND model* AND control* AND autonomous AND vehicle ) AND ( LIMIT-TO ( LANGUAGE, \"English\" ) ) AND ( LIMIT-TO ( PUBSTAGE, \"final\" ) ) AND ( LIMIT-TO ( DOCTYPE, \"ar\" ) ) AND ( LIMIT-TO ( PUBYEAR, 2020 ) OR LIMIT- TO ( PUBYEAR, 2021 ) OR LIMIT-TO ( PUBYEAR, 2022 ) OR LIMIT- TO ( PUBYEAR, 2023 ) OR LIMIT-TO ( PUBYEAR, 2024 ) )\u003c/p\u003e \u003cp\u003eDate of Access: October 2024\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eWoS\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\"system identification\" AND model* AND control* AND autonomous AND vehicle (Topic) and 2024 or 2023 or 2022 or 2021 or 2020 or 2019 (Publicati on Years) and Article (Document Types)\u003c/p\u003e \u003cp\u003e\u003cb\u003eDate of Access: October 2024\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Screening\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eIn the screening phase, each research item was carefully checked to ensure alignment with the defined research questions. This process is zero in studies pertinent to system identification for modeling and control within autonomous vehicles. The duplicate entries were systematically removed. From the initial pool, 235 studies were excluded, narrowing the selection to 83 papers for a deeper evaluation based on strict inclusion and exclusion criteria as shown in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. The review concentrated on English-language journal articles published between 2019 and 2024, deliberately setting aside conference papers, books, reviews, and any work still pending publication. Finally, 25 duplicates were excluded.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe selection criterion is searching\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCriterion\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eInclusion\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eExclusion\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eLanguage\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEnglish\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNon-English\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eTimeline\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2019\u0026ndash;2024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;2019\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eLiterature type\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eJournal (Article)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eConference, Book, Review\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003ePublication Stage\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFinal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eIn Press\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Eligibility\u003c/h2\u003e \u003cp\u003eDuring the eligibility phase, the third step (58 items) was chosen for an additional examination. At this juncture, the titles and principal content of each article were meticulously assessed to confirm their compliance with the inclusion criteria and alignment with the research objectives. Consequently, 27 papers were removed because they were outside the scope, possessed irrelevant titles, or presented abstracts unrelated to the objectives of the study. This procedure resulted in 31 papers for the final evaluation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Data Abstraction and Analysis\u003c/h2\u003e \u003cp\u003eThis study evaluated and synthesized several quantitative research designs using an integrated analysis. The main objective was to identify pertinent themes and subthemes aligned with the study subjects. The first phase involved collecting relevant data. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, the authors systematically reviewed 31 papers to extract critical information relevant to the research focus on system identification for modeling and control in autonomous vehicles, as listed in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. To ensure comprehensive coverage, this study assessed the methodologies and findings of multiple investigations. The process of refining themes required collaboration among the authors, as each theme was developed based on data collected within the study environment. A detailed notebook was maintained throughout the process of documenting reflections, interpretations, challenges, and evolving ideas during the data analysis. In the final phase, the authors cross-checked their findings to resolve any inconsistencies, and any disagreements regarding the themes were addressed through team discussions. The selected themes were carefully refined to ensure analytical consistency. This study addressed two key research questions:\u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eHow can advanced navigation-control strategies improve the operational efficiency and safety of autonomous vehicles in dynamic environments?\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eWhat role does machine learning play in enhancing the adaptability and robustness of control systems in autonomous vehicle applications?\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe primary data\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAuthors (Year)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTitle\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYear\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSource title\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eScopus\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eWoS\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eJiang et al. (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePath-following control of autonomous ground vehicles based on input convex neural networks\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2022\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eProceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAriza Ramirez et al. (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eExploration of the applicability of probabilistic inference for learning control in underactuated autonomous underwater vehicles\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2020\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAutonomous Robots\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWang et al. (\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eReal-time parameter identification of ship maneuvering response model based on nonlinear Gaussian Filter\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2022\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eOcean Engineering\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eKim et al. (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2024\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eField experiment of autonomous ship navigation in canal and surrounding nearshore environments\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eJournal of Field Robotics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWiberg et al. (\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2024\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSim-to-real transfer of active suspension control using deep reinforcement learning\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eRobotics and Autonomous Systems\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHumphreys et al. (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAdvancing Fusion with Machine Learning Research Needs Workshop Report\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2020\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eJournal of Fusion Energy\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIdros et al. (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2024\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eModelling of Vehicle Longitudinal Dynamics for Speed Control\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eJournal of Advanced Research in Applied Mechanics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMcCrink \u0026amp; Gregory (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDesign and Development of a High-Speed UAS for Beyond Visual Line-of-Sight Operations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2021\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eJournal of Intelligent and Robotic Systems: Theory and Applications\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRosolia \u0026amp; Borrelli (\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLearning How to Autonomously Race a Car: A Predictive Control Approach\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2020\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eIEEE Transactions on Control Systems Technology\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAhmed et al. (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2023\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSurvey on traditional and AI based estimation techniques for hydrodynamic coefficients of autonomous underwater vehicle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2023\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eOcean Engineering\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"No\" id=\"Taba\" border=\"1\"\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAuthors (Year)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTitle\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYear\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSource title\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eScopus\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eWoS\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTischler (\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eFlight Control Technology Advancements and Challenges for Future Rotorcraft 40th Alexander\u003c/p\u003e \u003cp\u003eA. Nikolsky Honorary Lecture\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2022\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eJournal of the American Helicopter Society\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWu et al. (\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSystem identification and controller design of a novel autonomous underwater vehicle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2021\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMachines\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOzdogan \u0026amp; Leblebicioglu (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDesign, Modeling, and Control Allocation of a Heavy-Lift Aerial Vehicle Consisting of Large Fixed Rotors and Small Tiltrotors\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2022\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eIEEE/ASME\u003c/p\u003e \u003cp\u003eTransactions on Mechatronics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eKawamura et al. (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eDevelopment of small autonomous surface vehicle implementing position control system using sliding mode control\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2021\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSensors and Materials\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLai \u0026amp; Le (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAdaptive Learning-Based Observer with Dynamic Inversion for the Autonomous Flight of an Unmanned Helicopter\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2021\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eIEEE Transactions on Aerospace and Electronic Systems\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMladenov \u0026amp; Best (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2023\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCharacterisation of driver longitudinal behaviour using an Unscented Kalman Filter\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2023\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eProceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eChen \u0026amp; Ozay (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eData-Driven Computation of Robust Control Invariant Sets With Concurrent Model Selection\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2022\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eIEEE Transactions on Control Systems Technology\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYan et al. (\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2024\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eIntegration of Vehicle Dynamic Model and System Identification Model for Extending the Navigation Service under Sensor Failures\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eIEEE Transactions on Intelligent Vehicles\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMartinsen et al. (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eReinforcement learning-based NMPC for tracking control of ASVs: Theory and experiments\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2022\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eControl Engineering Practice\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLee et al. (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2024\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNonlinear Model Predictive Control with Obstacle Avoidance Constraints for Autonomous Navigation in a Canal Environment\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eIEEE/ASME\u003c/p\u003e \u003cp\u003eTransactions on Mechatronics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSoitinaho \u0026amp; Oksanen (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2023\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLocal Navigation and Obstacle Avoidance for an Agricultural Tractor With Nonlinear Model Predictive Control\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2023\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eIEEE Transactions on Control Systems Technology\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"No\" id=\"Tabb\" border=\"1\"\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNo\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAuthors (Year)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eTitle\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eYear\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eSource title\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eScopus\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eWoS\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIvler et al. (\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSystem identification of rover dynamics: a comparison of three model structures\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2022\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eInternational Journal of Modelling, Identification and Control\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e23\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eT.K et al. (2020)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSystem identification of flybar-less rotorcraft UAV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2020\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAircraft Engineering and Aerospace Technology\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSouza et al. (\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2023\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eA Convolutional System Identification Approach Mixing Optimal Parameter Estimation and Deep Learning\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2023\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eInternational Journal of Control, Automation and Systems\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVicente et al. (\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLinear System Identification Versus Physical Modeling of Lateral-Longitudinal Vehicle Dynamics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2021\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eIEEE Transactions on Control Systems Technology\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDubey et al. (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2022\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSteering model identification and control design of autonomous ship: a complete experimental study\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2022\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eShips and Offshore Structures\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e27\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGrip et al. (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eModeling and identification of hover flight dynamics for NASA's Mars helicopter\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2020\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eJournal of Guidance, Control, and Dynamics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEsmaeili \u0026amp; Modares (\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2024\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRisk-Informed Model-Free Safe Control of Linear Parameter- Varying Systems\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eIEEE/CAA Journal of Automatica Sinica\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDeng et al. (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2021\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eIdentification of an Autonomous Underwater Vehicle hydrodynamic model using three Kalman filters\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2021\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eOcean Engineering\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGokul Krishnan et al. (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2024\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eReal-Time Experimental Evaluation and Analysis of PID and MPC Controllers Using HIL Setup for Robust Steering System of Autonomous Vehicles\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eIEEE Access\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e31\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMa et al. (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2023\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLocal Learning Enabled Iterative Linear Quadratic Regulator for Constrained Trajectory Planning\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2023\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eIEEE Transactions on Neural Networks and Learning Systems\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e/\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4. Result and Finding","content":"\u003cp\u003eThe study\u0026rsquo;s results and findings were organized into three primary themes. The initial theme, autonomous vehicles and navigation control, delves into the latest developments in real-time navigation strategies, path planning, and vehicle autonomy. The second theme, system identification and modeling techniques, emphasizes the importance of parameter estimation and model validation in the development of precise models of dynamic systems. The final theme, machine-learning and advanced control approaches, investigates the integration of data-driven algorithms with control systems to improve performance, adaptability, and decision making. These themes collectively offer a comprehensive perspective on the study's primary areas, emphasizing significant trends and insights into contemporary control engineering.\u003c/p\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e4.1 Autonomous Vehicles and Navigation Control\u003c/h2\u003e \u003cp\u003eVarious novel approaches and bodies of work for system identification are evident in the literature geared toward modeling and performing control of autonomous vehicles with different levels of autonomy ranging from human-run vehicle applications, such as agricultural tractors, autonomous surface vehicles (ASVs), and underwater vehicles. Considerable attention has been paid to the applications of nonlinear model predictive control (NMPC) and sliding mode control, which are both effective for navigation and obstacle avoidance. Soitinaho and Oksanen (\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) demonstrated how NMPC could be used in agricultural tractors, for example, when performing path tracking and obstacle avoidance. The experiments indicated that the NMPC technique maintains a cross-track error (xte) under 0.05 meters, demonstrating its physical world operating relevance. Similarly, Lee et al. (\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) used NMPC to improve the navigation performance of a small cruise boat operating in a canal-structured environment. Using experimental data and system identification, their setup validated NMPC for dealing with nonlinear dynamics and real-time LiDAR-induced obstacle avoidance. These results underscore the need for advanced control strategies to improve navigation systems in autonomous vehicles.\u003c/p\u003e \u003cp\u003eAn additional promising approach is the establishment of ASVs, developed by Kawamura et al. (\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). The authors proposed a sliding mode control system with robust properties against model mismatch and external disturbances. The ASV successfully navigated with minimal human supervision, indicating that the system could be trusted in perpetuating aquatic environments. Correspondingly, Kim et al. (\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) reported a study in which ASV autonomous capabilities were enhanced using additional sensors and control algorithms. The results from the field experiment showed that the system was able to sense its environment and autonomously make complex movements around obstacles, reaffirming sliding mode control as a practical controller in marine applications. Ariza Ramirez et al. (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) studied the use of probabilistic inference learning for autonomous underwater vehicles (AUVs) control. Conventional methods of controller tuning that gain some win in stationary backgrounds lose control when conditions are dynamic and practical, as mentioned in their study. Instead, they proposed a probabilistic model-based reinforcement learning procedure that can be adapted to different missions with almost no pre-programming and shows good promise for autonomous navigation of AUVs in uncertain environments. This drastically improves operational efficiency because fewer field trials are needed to find good navigation policies.\u003c/p\u003e \u003cp\u003eThe other is the ongoing development of flight control technologies for rotorcraft. In his review of the flight control system (Tischler, \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2022\u003c/span\u003e); he illustrated the help of time-domain and frequency-domain analyses to attain stability and handling quality for autonomous flight. Advanced feedback control systems have become essential to consistently meet the performance objectives of many rotorcraft applications such as urban air mobility. These developments illustrate the continued need for system identification and adaptive control to address the problems caused by new classes of vehicles. While modeling such systems can be challenging owing to the complexity of vehicle dynamics and motion, their importance as reliable control systems has sparked increasing interest in the system identification literature around AVs. Therefore, a potential path is derived from neural network design, namely input convex neural networks (ICNNs), which can provide unique benefits for certain path-following control tasks. Traditional modeling methods for AGVs are based on first principles are often impractical, which motivates a shift from model-based approaches to data-driven methods (Jiang et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The use of the properties of ICNNs transforms the predictive control problem into a convex optimization problem, thus allowing us to compute feasible solutions. An online learning algorithm periodically enhances the model to adapt it to various road conditions and disturbances, showing its effectiveness in simulations.\u003c/p\u003e \u003cp\u003eOne of the critical components of the system identification problem is the difficulty in identifying the steering dynamics between autonomous ships. An approach of system identification using input-output for a steering model was developed by Dubey et al. (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), which was used to identify the models for autonomous vessels. They then adopted models of similar traditional hydrodynamic modeling that lacked proper knowledge, which made it difficult to create efficient control systems. To investigate the behavior of vessels, they derived an autonomous ship model that has self-propelled autonomous capabilities and is equipped with IoT features to allow real-time data acquisition under different conditions. The experimental results highlight the viability of using IoT for hydrodynamic experiments and demonstrate that making data available can help improve control system performance. To enhance the accuracy of system identification techniques for AVs, a combination of state-of-the-art computational methods and real-time data access is required. Martinsen et al. (\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) and Jiang et al. (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) highlight that machine learning to privilege control strategy optimization will require higher degrees of manifestation, which is expected with the technological advancement cycle. Integrating methodologies from classical control theory, such as optimal and model predictive control upper bounds on task performance, with current AI design approaches (reinforcement learning and neural networks) will ultimately lead to autonomous vehicles that learn new skills while still conforming to their original signal bounds. Furthermore, Dubey et al. (\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) highlighted this gap and called toward more realistic experimental setups that help design better control systems in real-world autonomous maritime applications.\u003c/p\u003e \u003cp\u003eA literature review on system identification for autonomous vehicles shows significant contributions associated with computational models in terms of increasing the accuracy of modelled data and nonlinear behavior of control objects. Using machine learning approaches to identify systems has great potential for achieving the best performance and safety in AVs. Nonlinear model predictive control, sliding mode control, and learning-based approaches are state-of-the-art methodologies that aim to improve the navigation and control functionalities of autonomous vehicles. This highlights the need for significant system identification, testing, and control approaches to adapt to unknown or less than perfect scenarios with environmental dynamics. Further research should be conducted to continue developing these methodologies and applying them to other vehicle types and operational scenarios to develop safer and more efficient autonomous systems. With the development of autonomous technologies, such a model can also lead to innovative solutions that can deal with problems arising from newly encountered scenarios.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e4.2 System Identification and Modeling Techniques\u003c/h2\u003e \u003cp\u003eSystem identification is an important step in the modeling and control of autonomous vehicles, paving the way for the development of predictive models that are essential for control and navigation. The importance of this differential activation and contributions from multiple methodologies in the literature emphasize the complexity of system performance across diverse platforms and environments through effective responses. A rapid yet realistic mathematical model for vehicle longitudinal dynamics is essential for the speed control task and was highlighted by Idros et al. (\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). The detailed model proposed in this work is integrated with the dynamics of the vehicle body, powertrain, and braking, and simulation was performed using MATLAB Simulink to analyze various control strategies. The results reveal that the hierarchical PID control structure is capable of accurately tracking conventional urban drive cycles, thus validating the model as a platform for further control strategy development. For example, in autonomous navigation, to achieve better model fidelity, Wang et al. (\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) presented a method based on a nonlinear Gaussian filter for real-time parameter identification that can be utilized to adjust ship maneuvering response models. This investigation illustrates how the addition of observers can reduce parameter drift, leading to enhanced system identification performance. In contrast, these include the analysis of three model structures for rover dynamics and are designed with a higher fidelity model, reaching the conclusion that roll-yaw is the most accurate yet most complex of all others, whereas simpler models, such as dynamic bicycle models, cannot have the totality of other simplified characterizations (Ivler et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Together, these studies communicate the benefit of an appropriate model structure, given that one has estimated prediction error and must detect a balance between overfitting (by using complex models) and underfitting in dynamic landscapes.\u003c/p\u003e \u003cp\u003eIn this area, the literature mirrors the advancements in data-driven methods for system identification. Chen and Ozay (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) proposed an approach for synthesizing robust control-invariant sets and highlighted the link between model selection and optimal system identification from data. Through their approach, they showed that you can significantly improve the estimates of what makes control systems robust to maintain performance in uncertain environments. Similarly, Vicente et al. (\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) compared the lateral and longitudinal vehicle dynamics linear system identification with physical modeling. In addition, they concluded that data-driven models can outperform traditional physical models, at least when it comes to following real-world driving behavior, leading them away from more constrained methods of modeling autonomous vehicle systems. Deng et al. (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) explored this line of research further. This study employs different Kalman filter algorithms to identify hydrodynamic models for autonomous underwater vehicles and shows that optimized methods, such as the optimized unscented Kalman filter, can help improve the parameter estimation accuracy in noise. This finding is consistent with those of Yan et al. (\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Finally, they proposed a sensor-free localization approach combining system identification with vehicle dynamic models. These experiments show that not only is the localization accuracy greatly improved (especially during sensor failure scenarios) but also highlighted the capability of system identification to improve resilience in autonomous systems.\u003c/p\u003e \u003cp\u003eFinally, Ma et al. (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) used iterative and machine-learning techniques. However, this is the only example of a local learning-enhanced iterative linear quadratic regulator for trajectory planning. This demonstrates the potential of tighter coupling between system identification and model predictive control frameworks to achieve more efficient and flexible trajectory generation approaches for robotic and autonomous systems. Grip et al. (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) used a different modeling approach and dealt with challenges due to the Martian environment for the dynamic model of NASA's Mars helicopter and showed that proper system identification techniques are crucial when flying under extreme conditions. Overall, the literature on system identification for autonomous vehicles indicates a shift toward fusing different modeling methodologies to achieve more accurate and robust models while accounting for rich dynamic behavior. This shift toward data-driven approaches represents a larger trend in the literature in which experimental data are used to enhance model fidelity and control performance, making such systems more viable for autonomous navigation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e4.3 Machine-Learning and Advanced Control Approaches\u003c/h2\u003e \u003cp\u003eThe introduction of machine learning and advanced control approaches has led to recent advances in autonomous vehicles. With further emphasis on autonomous systems that need to be reliable and efficient, researchers have also stressed the best design and control methods. For example, Ozdogan and Leblebicioglu (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) recently introduced a new heavy-lift aerial vehicle (HLAV), which is designed with a special configuration of propellers that improves its controllability and requires no use of complex mechanisms. Their study emphasized global second-order sliding mode-based nonlinear dynamic modeling, where closed-loop system identification allows for the estimation of parameters and ultimately results in good trajectory tracking performance. Likewise, other researchers such as Gokul Krishnan et al. (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) tackled the problem of steering control systems in autonomous vehicles and compared two traditional controllers, PD and PID, with a recent Model Predictive Control (MPC) formulation. Through this study, they emphasized that real-time testing in diverse driving scenarios is vital for assessing the survivability of the designed controller for safe navigation.\u003c/p\u003e \u003cp\u003eIn addition, advanced deep learning techniques have been integrated into the system identification area for modeling dynamic systems using a convolutional system identification method (Souza et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). This hybrid approach results in the identification of complex behaviors that allow for more realistic scenarios than conventional models, thereby increasing the reliability of autonomous vehicle controls. Lai and Le (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) proposed an adaptive learning-based observer with dynamic inversion for flight control of an unmanned helicopter, where a control subject can utilize adaptive learning mechanisms to enhance the performance of control in a non-static environment. Machine-learning techniques have also been employed to model driver behavior. However, for examples closely related to longitudinal driver behavior characterization in real time, Mladenov and Best (\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) employed an Unscented Kalman Filter (UKF). Our method not only provides estimates of driver response parameters, but can also be used for insurance practice and autonomous driving systems to simulate different driving styles. Exploration of machine learning methods in driver behavior finds this interdisciplinary nature of advancements in autonomous vehicles where exploratory insights from control theory, data analysis, and behavioral modeling are used to predict the risk associated with driving.\u003c/p\u003e \u003cp\u003eFinally, we discuss how machine learning can be used in wider areas, such as fusion energy research, as a new tool. Humphreys et al. (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) addressed Fusion Energy Challenges With ML/AI Integration: Bridging Theory and Practice with Data-Driven Approaches. These multidisciplinary efforts continue to highlight the strides in tuning control and emphasize the inclusion of machine learning for better accuracy in autonomous systems. This journal focuses on advancing the dynamic interplay between controls, systems, and machines. System identification methods have been widely applied to the control of autonomous vehicles, particularly in machine learning and nonlinear model predictive control. One specific example is the development of risk-informed model-free control strategies (Esmaeili \u0026amp; Modares, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). They proposed a probabilistic safe control design for linear parameter-varying (LPV) systems that represents safety given closed-loop data and avoids model details. They demonstrated that the variance of a closed-loop system and the probability of safety are both inherently tied to the decision variables and noise covariances. This novel framework of risk-averse control design approaches safety by reducing the probability of safety violations with a proper data- driven control analysis. These safe control systems are necessary to provide solutions that account for the uncertainty in controlling autonomous vehicles.\u003c/p\u003e \u003cp\u003eAn additional major effort in this area comes from Wiberg et al. (\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) on heavy vehicles with active suspension systems using deep reinforcement learning (DRL) controllers and studying the sim-to-real transfer of DRL controllers. Unlike traditional robotic work that emphasizes lighter robots, their study faces unique challenges, such as the complicated behavior of a hydraulic driveline and slow actuation. Using system identification and sim-to-real mitigation techniques, such as domain randomization and action delays, the sim-to-real gap can be closed by fine-tuning the simulation parameters. Their findings suggest that training policies can be designed to provide not only simulation performance but also fidelity in real-world motion trajectories. This highlights the need for reliable modeling and simulation methods to obtain reliable control strategies for autonomous systems in difficult environments. Ahmed et al. (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) focused on determining the hydrodynamic coefficients of an autonomous underwater vehicle (AUV). Such a thorough survey is relevant and emphasizes that robust dynamic modeling will always be an important prerequisite for designing informative control systems in the case of AUVs moving in six degrees of freedom. This study compares traditional estimation methods (Analytical and Semi-Empirical approaches and CFD) to more recent artificial intelligence techniques (i.e., supervised machine-learning algorithms). The authors discussed the benefits of these newer methods, which are more accurate and computationally cheaper than older generation techniques. In addition, it explains the development of estimation methods in the last 75 years and their relevance in efficiently designing AUV control systems. Researchers have provided important insights into the enhancement of control strategies for autonomous marine vehicles by combining traditional methods with modern artificial intelligence techniques.\u003c/p\u003e \u003cp\u003eA strong interaction is identified between model-driven and machine-learning algorithms to ensure system reliability, performance, and adaptability in the ever-changing environment of autonomous cars, as presented in the literature. Such integration is indispensable for making future autonomous systems safe and efficient. From designs of safe control that are informed by risk, strategies around sim-to-real transfer as well as the notions of AI-based estimation methods being proposed collectively in multiple papers reflect a realization of the complex behavior these autonomous vehicles exhibit over their dynamics. Hence, these studies lay the groundwork for innovations to improve the effectiveness and safety of autonomous systems both on land and in water.\u003c/p\u003e \u003c/div\u003e"},{"header":"5. Discussion and Conclusion","content":"\u003cp\u003ePrior work on system identification for autonomous vehicles shows considerable documented development of control strategies for different vehicle types, such as agricultural, marine, and aerial systems. In particular, we have struggled with nonlinear model predictive control (NMPC) and sliding mode control to enhance navigation and obstacle avoidance. In addition, for land and water vehicles, namely NMPC, it is very helpful to tackle the path-tracking problem and environmental disturbances. Sliding mode control ensures the reliability of autonomous surface vehicles (ASVs) owing to their robustness to uncertainties and external disturbances in the environment. Probabilistic learning approaches are adaptable to variations in environments and can generalize their performance from a few field trials, which makes them appropriate for use with autonomous underwater vehicles (AUVs). In state-of-the-art aerial applications, flight control systems have been designed for frequency-domain analysis and adaptive feedback, with a view toward evolving solutions, such as urban air mobility rotorcraft.\u003c/p\u003e \u003cp\u003eBy using data-driven methods to analyze the results, this study simplifies traditional modeling approaches. Adaptive path-following and handling arbitrary richness scenarios are strongly enabled by Input Convex Neural Networks (ICNNs) with reinforcement-learning algorithms. They offer the advantages of design with models, where IoT-enabling capabilities, help enable data acquisition as required in real time, assisting strong hydrodynamic modeling and efficiency with control systems that are assisted by this. In particular, Advanced Kalman filters for AUV path recovery can also be employed as a method for parameter estimation in noise. Highly specific modeling, such as that for NASA\u0026rsquo;s Mars helicopter, shows that being able to identify the system precisely is key to performance in very hostile environments. This change to data-centric models is a response to the tendency of machine-learning methods to supersede physical models, increasing adaptability and robustness in systems that operate under dynamic and uncertain conditions.\u003c/p\u003e \u003cp\u003eThese continuing developments demonstrate the cooperation between machine learning and classic control approaches using resilient adaptive autonomous systems. Closed-loop system identification of a heavy-lift aerial vehicle and nonlinear model advantage for trajectory tracking. Comparative analyses between classical PID controllers and contemporary Model Predictive Control (MPC) highlight the importance of in-situ testing to validate safety and performance during navigation. Driver behavior modeling based on methods using machine learning, such as the Unscented Kalman Filter (UKF), has also found practical use in areas such as insurance assessment and autonomous systems. For this purpose, risk-aware control strategies have recently gained considerable attention to deal with uncertainties and, at the same time, mimicking real-world performance by sim-to-real transfer methods to boost simulation issues in such areas as domain randomization or action delay optimization, contributing to increasing confidence in heavy vehicle control performance focus areas.\u003c/p\u003e \u003cp\u003eFuture work should focus on developing these methods further and applying them more broadly to a larger class of vehicles and operating conditions as control strategies continue to advance. This will foster the development of secure and trustworthy autonomous systems capable of adapting to complex environments, while driving innovation in control and navigation. The literature indicates a prominent path for merging conventional control theories with data-driven methods and improving the autonomy of a system with higher robustness, flexibility, and efficacy. This evolution has enabled stronger navigation on different platforms and a stable environment in dynamic settings. By integrating conventional control methods with modern AI techniques, this approach provides a robust design framework for enhancing efficiency, safety, and adaptability of autonomous systems, whether land-, air-, or sea-based.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eFunding Statement:\u003c/h2\u003e \u003cp\u003eThis study was conducted as part of a personal interest and was not supported by any specific grant. The research aligns with the interests of the Dynamics Modeling \u0026amp; Control System (DyCoS) research group at Universiti Malaysia Perlis (UniMAP).\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eEthical Compliance\u003c/strong\u003e \u003cp\u003eThis study did not involve any human participants, animals, or other living subjects. It is a recent review focused on system identification methods for modeling autonomous vehicles.\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eConflict of Interest\u003c/strong\u003e \u003cp\u003e \u003cb\u003edeclaration\u003c/b\u003e: The authors declare that they have no affiliations with or involvement in any organization or entity with any financial interest in the subject matter or materials discussed in this manuscript.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eM.Z. Zakaria wrote the main manuscript text.M.S. Saad prepared Table 1-2.M.E. Baharudin prepared Figure 1.A. Mohd Nor prepared Table 3.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eAhmed, F., Xiang, X., Jiang, C., Xiang, G., \u0026amp; Yang, S. (2023). Survey on traditional and AI based estimation techniques for hydrodynamic coefficients of autonomous underwater vehicle. \u003cem\u003eOcean Engineering\u003c/em\u003e, \u003cem\u003e268\u003c/em\u003e. https://doi.org/10.1016/j.oceaneng.2022.113300\u003c/li\u003e\n \u003cli\u003eAriza Ramirez, W., Leong, Z. Q., Nguyen, H. D., \u0026amp; Jayasinghe, S. G. (2020). Exploration of the applicability of probabilistic inference for learning control in underactuated autonomous underwater vehicles. \u003cem\u003eAutonomous Robots\u003c/em\u003e, \u003cem\u003e44\u003c/em\u003e(6), 1121\u0026ndash;1134. https://doi.org/10.1007/s10514-020-09922-z\u003c/li\u003e\n \u003cli\u003eAziziaghdam, E. T., \u0026amp; Alankuş, O. (2022). Longitudinal Control of Autonomous Vehicles Consisting Power-Train With Non-Linear Characteristics. \u003cem\u003eIEEE Transactions on Intelligent Vehicles\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e, 133\u0026ndash;142. https://doi.org/10.1109/tiv.2021.3115265\u003c/li\u003e\n \u003cli\u003eBarrozo, J., \u0026amp; Lazcano, V. (2019). Simulation of an Autonomous Vehicle Control System Based on Image Processing. \u003cem\u003e2019 5th International Conference on Frontiers of Signal Processing (ICFSP)\u003c/em\u003e, 88\u0026ndash;94. https://doi.org/10.1109/icfsp48124.2019.8938094\u003c/li\u003e\n \u003cli\u003eBerrada, J., \u0026amp; Leurent, F. (2017). Modeling Transportation Systems involving Autonomous Vehicles: A State of the Art. \u003cem\u003eTransportation Research Procedia\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e, 215\u0026ndash;221. https://doi.org/10.1016/J.TRPRO.2017.12.077\u003c/li\u003e\n \u003cli\u003eChalvatzaras, A., Pratikakis, I., \u0026amp; Amanatiadis, A. (2023). A Survey on Map-Based Localization Techniques for Autonomous Vehicles. \u003cem\u003eIEEE Transactions on Intelligent Vehicles\u003c/em\u003e, \u003cem\u003e8\u003c/em\u003e, 1574\u0026ndash;1596. https://doi.org/10.1109/TIV.2022.3192102\u003c/li\u003e\n \u003cli\u003eChen, Y., \u0026amp; Ozay, N. (2022). Data-Driven Computation of Robust Control Invariant Sets With Concurrent Model Selection. \u003cem\u003eIEEE Transactions on Control Systems Technology\u003c/em\u003e, \u003cem\u003e30\u003c/em\u003e(2), 495\u0026ndash;506. https://doi.org/10.1109/TCST.2021.3069759\u003c/li\u003e\n \u003cli\u003eCulley, J., Garlick, S., Esteller, E. G., Georgiev, P., Fursa, I., Sluis, I. Vander, Ball, P., \u0026amp; Bradley, A. (2020). System Design for a Driverless Autonomous Racing Vehicle. \u003cem\u003e2020 12th International Symposium on Communication Systems, Networks and Digital Signal Processing (CSNDSP)\u003c/em\u003e, 1\u0026ndash;6. https://doi.org/10.1109/CSNDSP49049.2020.9249626\u003c/li\u003e\n \u003cli\u003eDasarathy, B. (1980). \u003cem\u003eInformation Processing For Target Recognition From Autonomous Vehicles\u003c/em\u003e. \u003cem\u003e0219\u003c/em\u003e. https://doi.org/10.1117/12.958535\u003c/li\u003e\n \u003cli\u003eDeng, F., Levi, C., Yin, H., \u0026amp; Duan, M. (2021). Identification of an Autonomous Underwater Vehicle hydrodynamic model using three Kalman filters. \u003cem\u003eOcean Engineering\u003c/em\u003e, \u003cem\u003e229\u003c/em\u003e. https://doi.org/10.1016/j.oceaneng.2021.108962\u003c/li\u003e\n \u003cli\u003eDubey, A. C., Subramanian, A. V, \u0026amp; Jagadeesh Kumar, V. (2022). Steering model identification and control design of autonomous ship: a complete experimental study. \u003cem\u003eShips and Offshore Structures\u003c/em\u003e, \u003cem\u003e17\u003c/em\u003e(5), 992\u0026ndash;1004. https://doi.org/10.1080/17445302.2021.1889193\u003c/li\u003e\n \u003cli\u003eEsmaeili, B., \u0026amp; Modares, H. (2024). Risk-Informed Model-Free Safe Control of Linear Parameter-Varying Systems.\u0026nbsp;\u003cem\u003eIEEE/CAA Journal of Automatica Sinica\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(9), 1918\u0026ndash;1932. https://doi.org/10.1109/JAS.2024.124479\u003c/li\u003e\n \u003cli\u003eF\u0026eacute;nyes, D., N\u0026eacute;meth, B., \u0026amp; G\u0026aacute;sp\u0026aacute;r, P. (2021). A Novel Data-Driven Modeling and Control Design Method for Autonomous Vehicles. \u003cem\u003eEnergies\u003c/em\u003e. https://doi.org/10.3390/EN14020517\u003c/li\u003e\n \u003cli\u003eGokul Krishnan, S., Suresh Kumar, P., Nassar Matara, N., \u0026amp; Wang, Y. (2024). Real-Time Experimental Evaluation and Analysis of PID and MPC Controllers Using HIL Setup for Robust Steering System of Autonomous Vehicles. \u003cem\u003eIEEE Access\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e, 74711\u0026ndash;74723. https://doi.org/10.1109/ACCESS.2024.3406219\u003c/li\u003e\n \u003cli\u003eGrip, H. F., Johnson, W., Malpica, C., Scharf, D. P., Mandić, M., Young, L., Allan, B., Mettler, B., Martin, M. S., \u0026amp; Lam, J. (2020). Modeling and identification of hover flight dynamics for NASA\u0026rsquo;s Mars helicopter. \u003cem\u003eJournal of Guidance, Control, and Dynamics\u003c/em\u003e, \u003cem\u003e43\u003c/em\u003e(2), 179\u0026ndash;194. https://doi.org/10.2514/1.G004228\u003c/li\u003e\n \u003cli\u003eHumphreys, D., Kupresanin, A., Boyer, M. D., Canik, J., Chang, C. S., Cyr, E. C., Granetz, R., Hittinger, J., Kolemen, E., Lawrence, E., Pascucci, V., Patra, A., \u0026amp; Schissel, D. (2020). Advancing Fusion with Machine Learning Research Needs Workshop Report. \u003cem\u003eJournal of Fusion Energy\u003c/em\u003e, \u003cem\u003e39\u003c/em\u003e(4), 123\u0026ndash;155. https://doi.org/10.1007/s10894-020-00258-1\u003c/li\u003e\n \u003cli\u003eIdros, S., Abdullah, M., Toha, S. F., \u0026amp; Said, M. S. M. (2024). Modelling of Vehicle Longitudinal Dynamics for Speed Control. \u003cem\u003eJournal of Advanced Research in Applied Mechanics\u003c/em\u003e, \u003cem\u003e123\u003c/em\u003e(1), 56\u0026ndash;74. https://doi.org/10.37934/ARAM.123.1.5674\u003c/li\u003e\n \u003cli\u003eIvler, C. M., Gowans, N., \u0026amp; Marfise, C. (2022). System identification of rover dynamics: a comparison of three model structures. \u003cem\u003eInternational Journal of Modelling, Identification and Control\u003c/em\u003e, \u003cem\u003e40\u003c/em\u003e(1), 70\u0026ndash;83. https://doi.org/10.1504/IJMIC.2022.124077\u003c/li\u003e\n \u003cli\u003eJames, S. S., Anderson, S., \u0026amp; Lio, M. (2020). Longitudinal Vehicle Dynamics: A Comparison of Physical and Data- Driven Models Under Large-Scale Real-World Driving Conditions. \u003cem\u003eIEEE Access\u003c/em\u003e, \u003cem\u003e8\u003c/em\u003e, 73714\u0026ndash;73729. https://doi.org/10.1109/ACCESS.2020.2988592\u003c/li\u003e\n \u003cli\u003eJiang, K., Hu, C., \u0026amp; Yan, F. (2022). Path-following control of autonomous ground vehicles based on input convex neural networks. \u003cem\u003eProceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering\u003c/em\u003e, \u003cem\u003e236\u003c/em\u003e(13), 2806\u0026ndash;2816. https://doi.org/10.1177/09544070221114690\u003c/li\u003e\n \u003cli\u003eKabzan, J., Hewing, L., Liniger, A., \u0026amp; Zeilinger, M. (2019). Learning-Based Model Predictive Control for Autonomous Racing. \u003cem\u003eIEEE Robotics and Automation Letters\u003c/em\u003e, \u003cem\u003e4\u003c/em\u003e, 3363\u0026ndash;3370. https://doi.org/10.1109/LRA.2019.2926677\u003c/li\u003e\n \u003cli\u003eKawamura, Y., Tahara, J., Kato, T., Fujii, S., Baba, S., \u0026amp; Koike, M. (2021). Development of small autonomous surface vehicle implementing position control system using sliding mode control. \u003cem\u003eSensors and Materials\u003c/em\u003e, \u003cem\u003e33\u003c/em\u003e(3), 883\u0026ndash;895. https://doi.org/10.18494/SAM.2021.3216\u003c/li\u003e\n \u003cli\u003eKim, J., Lee, C., Chung, D., Cho, Y., Kim, J., Jang, W., \u0026amp; Park, S. (2024). Field experiment of autonomous ship navigation in canal and surrounding nearshore environments. \u003cem\u003eJournal of Field Robotics\u003c/em\u003e, \u003cem\u003e41\u003c/em\u003e(2), 470\u0026ndash;489. https://doi.org/10.1002/rob.22262\u003c/li\u003e\n \u003cli\u003eLai, Y.-C., \u0026amp; Le, T.-Q. (2021). Adaptive Learning-Based Observer with Dynamic Inversion for the Autonomous Flight of an Unmanned Helicopter. \u003cem\u003eIEEE Transactions on Aerospace and Electronic Systems\u003c/em\u003e, \u003cem\u003e57\u003c/em\u003e(3), 1803\u0026ndash; 1814. https://doi.org/10.1109/TAES.2021.3050653\u003c/li\u003e\n \u003cli\u003eLee, C., Chung, D., Kim, J., \u0026amp; Kim, J. (2024). Nonlinear Model Predictive Control with Obstacle Avoidance Constraints for Autonomous Navigation in a Canal Environment. \u003cem\u003eIEEE/ASME Transactions on Mechatronics\u003c/em\u003e, \u003cem\u003e29\u003c/em\u003e(3), 1985\u0026ndash;1996. https://doi.org/10.1109/TMECH.2023.3312672\u003c/li\u003e\n \u003cli\u003eLi, B., Wang, R., Zhang, Y., \u0026amp; Wang, Z. (1999). Modeling of steering system of high speed intelligent vehicle by system identification. \u003cem\u003eProceedings of the IEEE International Vehicle Electronics Conference (IVEC\u0026rsquo;99) (Cat. No.99EX257)\u003c/em\u003e, 243\u0026ndash;246. https://doi.org/10.1109/IVEC.1999.830675\u003c/li\u003e\n \u003cli\u003eLi, L., Ota, K., \u0026amp; Dong, M. (2018). Humanlike Driving: Empirical Decision-Making System for Autonomous Vehicles. \u003cem\u003eIEEE Transactions on Vehicular Technology\u003c/em\u003e, \u003cem\u003e67\u003c/em\u003e, 6814\u0026ndash;6823. https://doi.org/10.1109/TVT.2018.2822762\u003c/li\u003e\n \u003cli\u003eLiniger, A., Domahidi, A., \u0026amp; Morari, M. (2015). Optimization‐based autonomous racing of 1:43 scale RC cars.\u0026nbsp;\u003cem\u003eOptimal Control Applications and Methods\u003c/em\u003e, \u003cem\u003e36\u003c/em\u003e, 628\u0026ndash;647. https://doi.org/10.1002/oca.2123\u003c/li\u003e\n \u003cli\u003eMa, J., Cheng, Z., Zhang, X., Lin, Z., Lewis, F. L., \u0026amp; Lee, T. H. (2023). Local Learning Enabled Iterative Linear Quadratic Regulator for Constrained Trajectory Planning. \u003cem\u003eIEEE Transactions on Neural Networks and Learning Systems\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e(9), 5354\u0026ndash;5365. https://doi.org/10.1109/TNNLS.2022.3165846\u003c/li\u003e\n \u003cli\u003eMartinsen, A. B., Lekkas, A. M., \u0026amp; Gros, S. (2022). Reinforcement learning-based NMPC for tracking control of ASVs: Theory and experiments. \u003cem\u003eControl Engineering Practice\u003c/em\u003e, \u003cem\u003e120\u003c/em\u003e. https://doi.org/10.1016/j.conengprac.2021.105024\u003c/li\u003e\n \u003cli\u003eMcCrink, M. H., \u0026amp; Gregory, J. W. (2021). Design and Development of a High-Speed UAS for beyond Visual Line- of-Sight Operations. \u003cem\u003eJournal of Intelligent and Robotic Systems: Theory and Applications\u003c/em\u003e, \u003cem\u003e101\u003c/em\u003e(2). https://doi.org/10.1007/s10846-020-01300-2\u003c/li\u003e\n \u003cli\u003eMercy, T., Parys, R. Van, \u0026amp; Pipeleers, G. (2018). Spline-Based Motion Planning for Autonomous Guided Vehicles in a Dynamic Environment. \u003cem\u003eIEEE Transactions on Control Systems Technology\u003c/em\u003e, \u003cem\u003e26\u003c/em\u003e, 2182\u0026ndash;2189. https://doi.org/10.1109/TCST.2017.2739706\u003c/li\u003e\n \u003cli\u003eMladenov, E., \u0026amp; Best, M. C. (2023). Characterisation of driver longitudinal behaviour using an Unscented Kalman Filter. \u003cem\u003eProceedings of the Institution of Mechanical Engineers, Part D: Journal of Automobile Engineering\u003c/em\u003e, \u003cem\u003e237\u003c/em\u003e(14), 3505\u0026ndash;3518. https://doi.org/10.1177/09544070221137402\u003c/li\u003e\n \u003cli\u003eMoher D, Liberati A, Tetzlaff J, A. D. (2009). PRISMA 2009 Flow Diagram. In \u003cem\u003eThe PRISMA statement\u0026nbsp;\u003c/em\u003e(Vol. 6, p.\u003c/li\u003e\n \u003cli\u003eMoriwaki, K., \u0026amp; Tanaka, K. (2007). Modelling and navigation control of autonomous vehicles. \u003cem\u003eSICE Annual Conference 2007\u003c/em\u003e, 427\u0026ndash;434. https://doi.org/10.1109/SICE.2007.4421022\u003c/li\u003e\n \u003cli\u003eNakamura, M., \u0026amp; Sakakibara, K. (2019). Formal Verification and Mathematical Optimization for Autonomous Vehicle Group Controllers. \u003cem\u003e2019 ACM/IEEE 22nd International Conference on Model Driven Engineering Languages and Systems Companion (MODELS-C)\u003c/em\u003e, 732\u0026ndash;733. https://doi.org/10.1109/MODELS-C.2019.00111\u003c/li\u003e\n \u003cli\u003eNamngam, K., Chitsakul, P., \u0026amp; Satthamsakul, S. (2022). Mathematical Model for an Autonomous Vehicle to Fulfill the Closed Area. \u003cem\u003e2022 6th International Conference on Imaging, Signal Processing and Communications (ICISPC)\u003c/em\u003e, 133\u0026ndash;138. https://doi.org/10.1109/ICISPC57208.2022.00032\u003c/li\u003e\n \u003cli\u003eOzdogan, G., \u0026amp; Leblebicioglu, K. (2022). Design, Modeling, and Control Allocation of a Heavy-Lift Aerial Vehicle Consisting of Large Fixed Rotors and Small Tiltrotors. \u003cem\u003eIEEE/ASME Transactions on Mechatronics\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e(5), 4011\u0026ndash;4021. https://doi.org/10.1109/TMECH.2022.3150713\u003c/li\u003e\n \u003cli\u003ePetrov, P., \u0026amp; Nashashibi, F. (2014). Modeling and Nonlinear Adaptive Control for Autonomous Vehicle Overtaking. \u003cem\u003eIEEE Transactions on Intelligent Transportation Systems\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e, 1643\u0026ndash;1656. https://doi.org/10.1109/TITS.2014.2303995\u003c/li\u003e\n \u003cli\u003eRosolia, U., \u0026amp; Borrelli, F. (2019). Learning How to Autonomously Race a Car: A Predictive Control Approach.\u003cem\u003eIEEE Transactions on Control Systems Technology\u003c/em\u003e, \u003cem\u003e28\u003c/em\u003e, 2713\u0026ndash;2719. https://doi.org/10.1109/TCST.2019.2948135\u003c/li\u003e\n \u003cli\u003eRosolia, U., \u0026amp; Borrelli, F. (2020). Learning How to Autonomously Race a Car: A Predictive Control Approach.\u0026nbsp;\u003cem\u003eIEEE Transactions on Control Systems Technology\u003c/em\u003e, \u003cem\u003e28\u003c/em\u003e(6), 2713\u0026ndash;2719. https://doi.org/10.1109/TCST.2019.2948135\u003c/li\u003e\n \u003cli\u003eRosolia, U., Carvalho, A., \u0026amp; Borrelli, F. (2016). Autonomous racing using learning Model Predictive Control. \u003cem\u003e2017 American Control Conference (ACC)\u003c/em\u003e, 5115\u0026ndash;5120. https://doi.org/10.23919/ACC.2017.7963748\u003c/li\u003e\n \u003cli\u003eSoitinaho, R., \u0026amp; Oksanen, T. (2023). Local Navigation and Obstacle Avoidance for an Agricultural Tractor With Nonlinear Model Predictive Control. \u003cem\u003eIEEE Transactions on Control Systems Technology\u003c/em\u003e, \u003cem\u003e31\u003c/em\u003e(5), 2043\u0026ndash;2054. https://doi.org/10.1109/TCST.2023.3291533\u003c/li\u003e\n \u003cli\u003eSouza, M. B. A., dos Santos Neto, A. F., de Mello Hon\u0026oacute;rio, L., de Oliveira, E. J., Silva, M. F., \u0026amp; Pancoti, A. (2023). A Convolutional System Identification Approach Mixing Optimal Parameter Estimation and Deep Learning. \u003cem\u003eInternational Journal of Control, Automation and Systems\u003c/em\u003e, \u003cem\u003e21\u003c/em\u003e(8), 2674\u0026ndash;2684. https://doi.org/10.1007/s12555- 022-0070-3\u003c/li\u003e\n \u003cli\u003eTischler, M. B. (2022). Flight Control Technology Advancements and Challenges for Future Rotorcraft 40th Alexander A. Nikolsky Honorary Lecture. \u003cem\u003eJournal of the American Helicopter Society\u003c/em\u003e, \u003cem\u003e67\u003c/em\u003e(4). https://doi.org/10.4050/JAHS.67.041001\u003c/li\u003e\n \u003cli\u003eT.K, K. N., V.P, L., \u0026amp; Singh, J. (2020). System identification of flybar-less rotorcraft UAV. \u003cem\u003eAircraft Engineering and Aerospace Technology\u003c/em\u003e, \u003cem\u003e92\u003c/em\u003e(10), 1483\u0026ndash;1493. https://doi.org/10.1108/AEAT-05-2019-0100\u003c/li\u003e\n \u003cli\u003eVargas, J., Alsweiss, S., Toker, O., Razdan, R., \u0026amp; Santos, J. (2021). An Overview of Autonomous Vehicles Sensors and Their Vulnerability to Weather Conditions. \u003cem\u003eSensors (Basel, Switzerland)\u003c/em\u003e, \u003cem\u003e21\u003c/em\u003e. https://doi.org/10.3390/s21165397\u003c/li\u003e\n \u003cli\u003eVicente, B. A. H., James, S. S., \u0026amp; Anderson, S. R. (2021). Linear System Identification Versus Physical Modeling of Lateral-Longitudinal Vehicle Dynamics. \u003cem\u003eIEEE Transactions on Control Systems Technology\u003c/em\u003e, \u003cem\u003e29\u003c/em\u003e(3), 1380\u0026ndash; 1387. https://doi.org/10.1109/TCST.2020.2994120\u003c/li\u003e\n \u003cli\u003eWang, S., Wang, L., Im, N., Zhang, W., \u0026amp; Li, X. (2022). Real-time parameter identification of ship maneuvering response model based on nonlinear Gaussian Filter. \u003cem\u003eOcean Engineering\u003c/em\u003e, \u003cem\u003e247\u003c/em\u003e. https://doi.org/10.1016/j.oceaneng.2021.110471\u003c/li\u003e\n \u003cli\u003eWiberg, V., Wallin, E., F\u0026auml;lldin, A., Semberg, T., Rossander, M., Wadbro, E., \u0026amp; Servin, M. (2024). Sim-to-real transfer of active suspension control using deep reinforcement learning. \u003cem\u003eRobotics and Autonomous Systems\u003c/em\u003e,\u0026nbsp;\u003cem\u003e179\u003c/em\u003e. https://doi.org/10.1016/j.robot.2024.104731\u003c/li\u003e\n \u003cli\u003eWu, B., Han, X., \u0026amp; Hui, N. (2021). System identification and controller design of a novel autonomous underwater vehicle. \u003cem\u003eMachines\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e(6). https://doi.org/10.3390/machines9060109\u003c/li\u003e\n \u003cli\u003eXu, D., Shi, Y., \u0026amp; Ji, Z. (2018). Model-Free Adaptive Discrete-Time Integral Sliding-Mode-Constrained-Control for Autonomous 4WMV Parking Systems. \u003cem\u003eIEEE Transactions on Industrial Electronics\u003c/em\u003e, \u003cem\u003e65\u003c/em\u003e, 834\u0026ndash;843. https://doi.org/10.1109/TIE.2017.2739680\u003c/li\u003e\n \u003cli\u003eXu, S., Zidek, R. A. E., Cao, Z., Lu, P., Wang, X., Li, B., \u0026amp; Peng, H. (2022). System and Experiments of Model- Driven Motion Planning and Control for Autonomous Vehicles. \u003cem\u003eIEEE Transactions on Systems, Man, and Cybernetics: Systems\u003c/em\u003e, \u003cem\u003e52\u003c/em\u003e, 5975\u0026ndash;5988. https://doi.org/10.1109/tsmc.2021.3131141\u003c/li\u003e\n \u003cli\u003eYan, P., Wen, W., \u0026amp; Hsu, L.-T. (2024). Integration of Vehicle Dynamic Model and System Identification Model for Extending the Navigation Service under Sensor Failures. \u003cem\u003eIEEE Transactions on Intelligent Vehicles\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e(1), 2236\u0026ndash;2248. https://doi.org/10.1109/TIV.2023.3273185\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"System Identification, Autonomous Vehicles, Mathematical Modeling, Machine Learning, Advanced Control, PRISMA","lastPublishedDoi":"10.21203/rs.3.rs-5730506/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5730506/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis systematic literature review aims to identify recent trends and developments in system identification for the modeling and control of autonomous vehicles. Self-driving cars require robust operational dynamics that require modeling to ensure that the vehicles perform complex tasks and respond to changes in the working environment. In response to this, efforts were made to follow the guidelines of the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) checklist. Following pilot testing and database selection, Scopus and Web of Science searches produced 31 primary studies that met the inclusion criteria. These studies are categorised into three themes: The special topics presented include: (1) Autonomous Vehicles and Navigation Control, consisting of recent developments in path planning, obstacle detection, and mode switching; (2) System Identification and Modeling Techniques, which discusses dynamic model identification, real-time parameter estimation, and observer-based methodology; and (3) Machine Learning and Advanced Control Approaches, which discusses the integration of data-driven models, reinforcement learning, and hybrid control systems on vehicles. The findings indicate that integrating conventional control theories with contemporary advanced machine learning reduces reliability, flexibility, and performance. They also highlight how AV should obtain real-time data and IoT to enhance the performance of the control system under conditions of uncertainty. Considering this, this review finds that system identification remains a fundamental area to make breakthroughs in the development of autonomous vehicles because it offers a link between simulation and real-world results. Therefore, the findings offer a guideline for future research focusing toward making control strategies more intelligent and robust with policies for safer and more efficient auto referent systems in land, airborne, and water vehicle systems.\u003c/p\u003e","manuscriptTitle":"A Recent Systematic Review: System Identification for Modeling and Control in Autonomous Vehicles","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-12-31 06:03:17","doi":"10.21203/rs.3.rs-5730506/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"17376473-fe70-4d2f-ab3d-06c06b8d4044","owner":[],"postedDate":"December 31st, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-12-31T06:03:20+00:00","versionOfRecord":[],"versionCreatedAt":"2024-12-31 06:03:17","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5730506","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5730506","identity":"rs-5730506","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}