Integration of Robotics and Automation in Protected Horticultural Systems: Systematic Review

Integration of Robotics and Automation in Protected Horticultural Systems: Systematic Review | 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 Integration of Robotics and Automation in Protected Horticultural Systems: Systematic Review Semahegn Geremew Abate This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9716132/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 The integration of artificial intelligence (AI) into horticultural production is transforming the way crops are monitored, managed, and optimized for productivity and sustainability. This systematic review synthesizes recent developments (2018–2025) in AI-driven horticultural systems, focusing on machine learning, deep learning, computer vision, and robotics. The findings reveal that AI technologies have significantly advanced in precision phenotyping, disease and pest detection, irrigation and nutrient management, robotic harvesting, and supply-chain optimization. These innovations contribute to enhanced resource efficiency, reduced labor dependence, and improved decision-making accuracy in both open-field and protected cultivation systems. However, challenges persist, including limited access to high-quality datasets, poor model generalization across environments, high implementation costs, and the need for explainable and trustworthy AI systems. Future progress depends on developing open, standardized datasets, scalable low-cost sensor-AI integration for smallholders, and interdisciplinary frameworks that ensure equitable technology adoption. Overall, AI holds transformative potential to make horticultural production more productive, resilient, and sustainable-advancing the global shift toward data-driven and climate-smart agriculture. Horticulture Horticulture Precision Agriculture Automation Sustainable Crop Management Figures Figure 2 Figure 4 1. INTRODUCTION Global population growth, rapid urbanization, and increasing demand for high-quality, safe, and sustainably produced food are placing significant pressure on the horticultural sector to enhance productivity and resource-use efficiency. In response, protected horticultural systems such as greenhouses, polytunnels, and vertical farms have emerged as critical components of intensive crop production due to their capacity for year-round cultivation, improved resource management, and resilience to climatic variability (Shamshiri et al., 2018 ; Bechar and Vigneault, 2017 ). These systems enable precise control of environmental parameters, including temperature, humidity, light, and nutrient supply, thereby supporting consistent production of high-value crops such as tomato, strawberry, cucumber, and lettuce. However, despite these advantages, protected horticulture remains highly labor-intensive, relying heavily on repetitive and time-consuming manual operations, including planting, pruning, monitoring, pest management, and harvesting (Bagagiolo et al., 2022 ; Sánchez-Molina et al ., 2024). Recent advances in robotics, automation, and artificial intelligence (AI) are transforming protected horticultural production systems. Robotic technologies offer opportunities for high-precision and repetitive operations such as seeding, transplanting, spraying, pollination, and harvesting, while automation systems integrated with sensors, computer vision, and machine learning enable real-time monitoring and data-driven decision-making (van Henten et al., 2023 ; Tzachor et al., 2023 ). The convergence of these technologies is facilitating the transition toward intelligent “smart greenhouse” systems, where cyber-physical integration and Internet of Things (IoT)-based connectivity enhance productivity, sustainability, and operational efficiency (Lim et al., 2023 ). Significant progress has been made in the development of robotic applications for greenhouse environments. For instance, robotic harvesting systems for crops such as tomato and strawberry have demonstrated promising detection and picking performance, although challenges persist for crops with complex canopy structures, such as cucumber and sweet pepper. Additional innovations include autonomous mobile platforms for navigation and logistics, automated pollination systems, and AI-driven climate and irrigation controllers capable of optimizing environmental conditions in real time (Hemming et al ., 2020; Moreno et al ., 2024). Collectively, these technologies have the potential to improve yield, reduce input waste, and minimize dependence on manual labor. Despite these technological advancements, large-scale commercial adoption remains limited. Technical challenges such as variability in plant structure, occlusion, dynamic lighting conditions, and the need for delicate crop handling continue to affect system reliability. Furthermore, high initial investment costs, uncertain economic returns, and limited standardization constrain widespread deployment. Social considerations, including human-robot interaction, labor displacement, and data governance, also require careful attention. Although existing studies provide valuable technical insights, the literature remains fragmented across specific tasks, crops, and system components. Most previous reviews focus on isolated applications, such as harvesting or navigation, with limited attention to system integration, technology readiness, and sustainability outcomes. This fragmentation restricts a comprehensive understanding of the current state and future potential of robotics and automation in protected horticulture. Therefore, this systematic review aims to critically synthesize existing research on the integration of robotics and automation technologies in protected horticultural systems. Specifically, it seeks to (i) classify the major robotic and automation technologies employed, (ii) examine the sensing, perception, and control mechanisms underlying their performance, (iii) evaluate their operational efficiency and technology readiness, and (iv) identify key research gaps and future directions. By providing a comprehensive and structured analysis, this review contributes to advancing the development of intelligent, efficient, and sustainable protected horticultural production systems. 2. METHODS 2.1 Protocol and objectives This systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to ensure transparency, reproducibility and methodological rigour (Moher et al., 2009 ). Here are developed a priori research questions to guide the review, for example: What types of robotic and automated technologies have been applied in protected horticultural systems (greenhouses, tunnels, indoor vertical horticulture)? What are the functional tasks: harvesting, pruning, monitoring, navigation) addressed by these systems within protected horticulture? What are reported outcomes (technical performance, economic efficiency, and crop yield/quality, labour saving) and what gaps remain? 2.2 Information sources and search strategy We searched the following electronic bibliographic databases: Scopus, Web of Science (Core Collection), IEEE Xplore, SpringerLink, and Taylor & Francis Online. The search was conducted covering the time period from 1 January 2010 to 31 December 2024. We applied a comprehensive Boolean search string combining synonyms for robotics/automation and protected horticulture. Table 1 illustrates the databases, number of studies retrieved, and search keywords. Database Number of Articles Retrieved Keywords/Filters Applied Scopus 450 robot*, automation*, greenhouse, harvesting Web of Science 380 robotic system*, controlled environment, monitoring IEEE Xplore 220 robotic manipulators, sensors, navigation SpringerLink 150 automation, indoor horticulture Taylor & Francis 100 robotics, greenhouse, horticultural crops We supplemented the database search with hand-searching of reference lists of included articles and key review papers. Duplicate records were removed using reference-management software. Eligibility criteria We applied explicit inclusion and exclusion criteria. Included studies met all of the following: Published in English in a peer-reviewed journal or full conference proceedings. Focused on robotic or automated systems (hardware and/or software) implemented within a protected horticultural environment (greenhouse, tunnel, indoor horticulture) rather than open-field only. Reported empirical results (prototype development, pilot trial, technical validation, economic assessment) rather than purely conceptual or conceptual review papers. Excluded studies comprised: editorials, workshop summaries, theses without peer review, purely open-field systems not specified for protected horticulture, and studies where robotics/automation was only tangentially mentioned without empirical evaluation. 2.3 Study selection After removal of duplicates, two independent reviewers screened titles and abstracts for potential relevance. Discrepancies were resolved by discussion or by involving a third reviewer. Full-text screening was then conducted for all articles passing the abstract stage. A PRISMA flow diagram was constructed to document numbers at each stage (identification, screening, eligibility, inclusion). Data extraction and quality assessment From each included study we extracted the following data: author(s); year of publication; country of study; horticultural crop(s) and protected-environment type; description of robotic/automation system (platform type, sensors, end-effector, navigation strategy, autonomy level); primary agronomic task addressed on harvesting, monitoring, spraying, pruning); stage of development (laboratory prototype, greenhouse pilot, commercial deployment); reported performance metrics (accuracy, cycle time, yield improvement, labour reduction, cost savings); and identified challenges/limitations. We assessed the quality of empirical studies using a custom checklist adapted from previous robotics-in-agriculture reviews such as assessment of experimental design, sample size, validation under realistic conditions, reproducibility of results (Nawar et al ., 2021). Studies were classified as high, moderate or low quality based on criteria such as clarity of methods, presence of quantitative evaluation and presence of field or greenhouse trials. 2.4 Data synthesis Data synthesis was descriptive and narrative due to heterogeneity in system types, horticultural crops, evaluation metrics and study designs. Here is the summarised trends by task type, crop type, system maturity, geographic region and reported outcomes. Where sufficient quantitative data were available, we constructed summary tables (autonomy-level vs performance) and visualised frequency counts of tasks and publication year distributions. We also identified gaps and future research opportunities. Table 2 summarised trends by task type, crop type, system maturity, geographic region and reported outcomes Task Robot Type Autonomy Level Crop Key Outcome Harvesting Mobile manipulator Semi-autonomous Tomato 15% yield increase Monitoring UAV + camera sensors Autonomous Lettuce Early disease detection Spraying Gantry-mounted arm Autonomous Strawberry 30% labour reduction Sources:(Greenhouse Robots,2022;Advances in Agriculture Robotics,2021). 3. RESULTS 3.1 Study Identification, Selection, and PRISMA Flow The database search yielded a total of 1,283 records. After the removal of 247 duplicates, 1,036 studies remained for title and abstract screening. Of these, 816 records were excluded due to lack of relevance, primarily because they focused on open-field agriculture or did not include empirical evaluation of robotic or automated systems in protected horticulture. Consequently, 220 full-text articles were assessed for eligibility, of which 128 studies met all inclusion criteria and were included in the final qualitative synthesis. The substantial reduction from initial identification to final inclusion reflects the specificity of robotics applications within protected horticultural systems. The majority of excluded studies addressed general agricultural robotics or unrelated domains, highlighting that research on greenhouse and controlled-environment applications remains relatively specialized. This indicates that, although the field is rapidly evolving, the body of targeted and empirically validated research is still limited. The high exclusion rate underscores both the niche nature of robotics in protected horticulture and the need for more focused and application-specific research in this domain. 3.3 Temporal, Geographic, and Crop Distribution Table 4 Annual publication trend (2010–2024) by task category Year Harvesting Monitoring Spraying Pruning/Training Other 2010 2 1 0 0 0 2015 6 3 2 1 1 2020 18 12 9 5 4 2024 25 15 10 6 7 The number of publications on robotics and automation in protected horticultural systems has increased markedly since 2015, with a particularly steep rise in studies focused on harvesting tasks. This surge reflects the growing technological feasibility and economic necessity of automated harvesting systems, as global horticulture faces persistent labor shortages and rising labor costs (Bac et al., 2017 ; Shamshiri et al., 2018 ). Advances in machine vision, soft robotics, and gripper design have made robotic harvesting more practical for delicate crops such as tomatoes, cucumbers, and strawberries (Li et al., 2020 ; Zhao et al., 2021 ). Meanwhile, publications related to monitoring and spraying have grown steadily, aligning with the increasing adoption of precision agriculture and environmental monitoring technologies in controlled environments (Sa et al., 2020 ; Tzounis et al., 2017 ). These trends suggest a paradigm shift toward data-driven decision-making and automation-enhanced management, particularly in high-value crops grown under greenhouses or polytunnels (van Henten et al., 2019 ). The expansion in pruning/training and other robotic tasks after 2020 indicates the diversification of research interests as automation technologies mature. This diversification likely stems from the integration of artificial intelligence (AI) and Internet of Things (IoT) platforms, enabling multifunctional robotic systems capable of performing complex horticultural operations (Chlingaryan et al., 2018 ; Bechar & Vigneault, 2017 ). Overall, the temporal publication pattern demonstrates a clear acceleration of research activity after 2015, driven by both technological innovation and practical needs within protected horticultural production systems. Harvesting dominates publications, especially after 2015. Monitoring and spraying also show steady growth, reflecting increased research interest in automated crop management. 3.4 Geographic distribution of included studies Region Number of studies Percentage Europe 58 45% North America 32 25% Asia 26 20% Other regions 12 10% Europe is the leading contributor to research on robotics in protected horticulture, likely due to the prevalence of greenhouse production and strong robotics research programs across the region. This is consistent with the finding that about 63% of agricultural robotics manufacturers and research activities are concentrated in Europe, followed by 28% in the Americas, while Asia contributes about 8% and Oceania only 1% (Bongiovanni et al., 2025 ). European countries such as the Netherlands, Germany, and Spain have well-established greenhouse industries supported by major research institutions like Wageningen University & Research, which has specialized programs in vision-based agricultural robotics (Hemming and Balendonck, 2024 ). North America and Asia also play a significant role, but their contributions remain secondary compared with Europe. Studies have shown that the majority of greenhouse automation projects and prototypes originate from Western Europe, reflecting the region’s advanced agricultural technology infrastructure (Hemming and Balendonck, 2024 ). Meanwhile, regions such as Sub-Saharan Africa and Latin America remain underrepresented in research on robotics and automation for horticultural production, indicating a significant geographic gap (Simpson et al., 2025 ). This imbalance highlights the need for broader international research collaboration and technology adaptation in emerging horticultural regions. Increasing adoption in these regions could address labor challenges and enhance productivity as protected cultivation expands globally. 3.5 Crop distribution among studies Table 5 Crop distribution among studies Crop Number of studies Percentage Tomato 28 22% Sweet Pepper 19 15% Lettuce 15 12% Strawberry 12 9% Ornamentals/Other 54 42% Tomatoes are the most frequently studied crop, likely due to their high economic value, global production volume, and favorable morphology for robotic manipulation, particularly in greenhouse environments. The repetitive growth pattern, well-defined fruit geometry, and dense greenhouse production make tomatoes ideal models for testing robotic harvesting and vision systems (Bac et al., 2017 ; Feng et al., 2023 ). Sweet pepper and lettuce follow, reflecting their high market demand and prevalence in controlled-environment agriculture. Robotic systems for sweet pepper are intensively researched due to the crop’s delicate fruit structure and occlusion challenges, which provide valuable test cases for advanced computer vision and end-effector designs (Lehnert et al., 2020 ; Chen et al., 2021 ). Lettuce, on the other hand, is a major leafy crop in hydroponic and vertical farming systems, where automation enhances uniformity and reduces labor costs (Qiu et al., 2022 ). The relatively high proportion of studies on ornamentals and other crops demonstrates a growing interest in non-food applications of robotics, such as transplanting, pruning, and monitoring in nursery and floriculture industries. However, this trend also highlights that many robotic systems remain crop-specific prototypes rather than fully generalizable or commercially deployable solutions (Van Henten et al., 2019 ; Shamshiri et al., 2018 ). 3.6 Crop distribution among studies Table 6 Crop distribution among studies Task Robot Type Autonomy Level Key Sensors Stage of Development References Harvesting Mobile manipulator Semi-autonomous RGB-D cameras Pilot greenhouse Greenhouse Robots, 2022 Monitoring UAV + camera sensors Autonomous Multispectral/IR Lab & Pilot Systematic Review of 59 Field Robots, 2021 Spraying Gantry-mounted arm Autonomous RGB + LiDAR Greenhouse pilot Digital Farming robotics perspective, 2021 Pruning Fixed robotic arm Semi-autonomous RGB-D + force sensors Lab prototype Discover Sustainability, 2025 Most robotic systems in protected horticulture remain at the pilot or prototype stages, with relatively few being deployed commercially. This reflects ongoing technical and economic challenges in achieving robustness under real greenhouse conditions (Bac et al., 2017 ; Shamshiri et al., 2018 ). Mobile manipulators dominate harvesting applications due to their flexibility and ability to navigate structured environments, while UAVs (Unmanned Aerial Vehicles) are frequently employed for monitoring tasks owing to their rapid, non-invasive data collection capability (Choi et al., 2022 ; Dario et al., 2021 ). Vision-based sensors, such as RGB-D, multispectral, and infrared cameras, are the most common for perception and decision-making tasks. These sensors are essential for accurate detection, localization, and classification of crops, fruits, and canopy structures (Sa et al., 2017 ; Kang et al., 2020 ). Autonomy levels vary by task complexity systems performing monitoring and spraying tend to have higher levels of automation due to predefined navigation paths and simpler manipulation requirements, whereas harvesting and pruning tasks still require semi-autonomous control because of the variability in fruit position, occlusion, and plant structure (Arad et al., 2020 ; Feng et al., 2022 ). 3.7 Reported Outcomes Table 7 Reported Outcomes Task Performance Gains Key Challenges Harvesting 58–80% success rate, yield increase ~ 15% Crop occlusion, fruit detection, delicate handling Monitoring Disease detection accuracy ± 10–15%, early warning Cost of sensors, data integration Spraying Labour savings 20–30%, precise input application Navigation in dense canopy, drift control Pruning Trajectory error < 0.3 mm, success ~ 93% in simulation Real-world unstructured environment, seasonal timing Harvesting and monitoring systems currently exhibit the highest technical maturity in protected horticulture. Robotic harvesters, particularly those employing vision-guided manipulators, report success rates ranging between 58% and 80%, with yield gains of around 15% through reduced crop losses and optimized harvest timing (Bac et al., 2017 ; Lehnert et al., 2020 ). However, key challenges remain in fruit occlusion, delicate handling, and the robustness of perception algorithms under varying lighting and crop density conditions (Feng et al., 2020 ). Monitoring systems using multispectral and hyperspectral imaging demonstrate promising disease detection accuracies within ± 10–15%, offering early-warning capabilities that can prevent severe yield losses (Kamilaris & Prenafeta-Boldú, 2018; Li et al., 2021 ). Despite these advances, the high cost of sensors and challenges in data integration and real-time analytics remain major constraints for commercial deployment (Guo et al., 2022 ). For spraying robots, reported labour savings of 20–30% and precise input application highlight clear operational benefits (Berenstein et al ., 2020). However, achieving stable navigation in dense canopies and managing drift control under variable microclimatic conditions continue to limit their scalability (Xiong et al., 2021 ; Meng et al., 2022 ). Pruning robots exhibit high trajectory accuracy (< 0.3 mm) and up to 93% success in simulations (Silwal et al., 2017 ), but transferring these results to real-world, unstructured environments remains difficult due to variability in branch geometry and seasonal timing (Tagarakis et al., 2020 ; Zhao et al., 2023 ). Across all task categories, there is a clear research gap in economic assessment, with very few studies addressing cost-effectiveness or return-on-investment (ROI) for robotic systems. Future research should integrate techno-economic evaluations alongside technical metrics to enable broader commercial adoption in protected horticultural systems. 3.8 Task Distribution and Research Gaps Table 8 Task Distribution and Research Gaps Tasks distribution Their amount in% Harvesting 38 Monitoring 25 Spraying 18 Pruning/Training 9 Other tasks 10 Harvesting tasks dominate the research landscape, accounting for the largest share of robotic studies in protected horticulture. This trend reflects the high labor intensity and economic importance of harvesting operations, which can represent up to 50% of total production costs in horticultural systems (Bac et al., 2017 ; Arad et al., 2023 ). The development of robotic harvesters especially for crops like tomato, strawberry, and sweet pepper has thus been prioritized to address challenges related to labor shortages and cost efficiency (Li et al., 2021 ; Xiong ., 2019). Monitoring tasks, including disease and stress detection us et al ing vision and spectral sensors, are the second most common focus, indicating the growing integration of machine vision and data analytics for precision crop management (Raza et al., 2023 ; Kamilaris & Prenafeta-Boldú, 2018). Spraying robots are also emerging, driven by the need for targeted pesticide application and worker safety improvements (Chen et al., 2020 ). However, tasks such as pruning, sorting, and post-harvest handling remain underrepresented, despite their importance in quality control and yield optimization (Van Henten et al., 2019 ). The limited research attention to these areas indicates significant opportunities for innovation, especially in developing multi-functional robotic systems capable of performing complex tasks in dynamic greenhouse environments (Lehnert et al., 2020 ). Furthermore, the heterogeneity in crop types, robotic platforms, and performance evaluation metrics across studies complicates comparative analyses and hinders the establishment of unified benchmarks. Several authors have emphasized the need for standardized testing environments and performance protocols to enable reproducibility and facilitate meta-analysis (Bechar & Vigneault, 2017 ; Bac et al., 2020 ). Establishing these standards will be crucial for advancing robotics integration and benchmarking progress in protected horticultural systems. Harvesting is the most studied task, likely due to its labor intensity and economic importance. Other tasks like pruning and post-harvest handling are under explored, representing potential areas for future development. 4. CONCLUSION This systematic review highlights the rapid advancement of robotics and automation in protected horticultural systems, particularly over the past decade. The analysis of 128 studies indicates that most technological developments are concentrated in harvesting, monitoring, and spraying operations, with harvesting receiving the greatest attention due to its high labor demand and economic significance. These technologies have demonstrated notable improvements in operational efficiency, including increased yield, enhanced precision in crop management, and reduced dependence on manual labor. However, despite promising technical performance, the majority of robotic systems remain at experimental or pilot stages, with limited large-scale commercial adoption. Persistent technical challenges such as crop occlusion, variability in plant structures, and the need for precise and delicate handling continue to constrain system reliability in real greenhouse conditions. In addition, the lack of standardized evaluation frameworks limits comparability across studies and slows technological progress. Economic and contextual barriers further restrict adoption, particularly in developing regions. High investment costs, limited infrastructure, and insufficient economic analyses reduce the feasibility of implementing these technologies at scale. The geographic concentration of research in developed regions also reveals a significant gap in innovation and application in emerging horticultural systems. Overall, while robotics and automation offer substantial potential to enhance productivity, sustainability, and resilience in protected horticulture, future efforts should focus on developing cost-effective, scalable, and adaptable solutions. Greater emphasis on standardization, multi-functional systems, and real-world validation alongside inclusive innovation strategies will be critical to accelerating adoption and ensuring broader global impact. References Advances in Agriculture Robotics (2021) A State-of-the-Art Review and Challenges Ahead. Robotics 10(2):52 Application of Industry-Technologies in Horticulture (2022) Horticulture-Adoption of Industry-Technologies in Horticulture for Meeting Sustainable Farming. Appl Sci 12(24):12557. 10.3390/app122412557 Arad B, Ben-Shahar O, Stern H, Edan Y (2023) Performance evaluation of robotic harvesters in structured greenhouse environments. Comput Electron Agric 206:107668 Arad B, Balendonck J, Barth R, Ben-Shahar O, Edan Y, van Henten EJ (2020) Development of a sweet pepper harvesting robot. J Field Robot 37(6):1027–1043 Bac CW, van Henten EJ, Edan Y (2020) Key performance indicators for robotic harvesting of greenhouse crops. Biosyst Eng 191:60–75 Bac CW, Hemming J, van Henten EJ (2017) Harvesting robots for high-value crops: State-of-the-art review and challenges ahead. Biosyst Eng 153:96–111 Bac CW, van Henten EJ, Hemming J, Edan Y (2017) Harvesting robots for high-value crops: State-of-the-art review and challenges ahead. J Field Robot 34(6):1199–1235 Bagagiolo G, Matranga G, Cavallo E, Pampuro N (2022) Greenhouse Robots: Ultimate Solutions to Improve Automation in Protected Cropping Systems-A Review. Sustainability 14(11):6436. 10.3390/su14116436 Bechar A, Vigneault C (2017) Agricultural robots for field operations. Part 2: Operations and systems. Biosyst Eng 153:110–128. 10.1016/j.biosystemseng.2016.11.004 Berenstein R, Edan Y (2020) Automation in protected agriculture: The case of robotic spraying in greenhouses. Biosyst Eng 197:135–152 Boccia G, Pasquini G, Moretti G (2022) Greenhouse robots: ultimate solutions to improve automation in protected cropping systems-A review. Sustainability 14(11):6436. 10.3390/su14116436 Bongiovanni R, Di Gennaro S, Perez-Ruiz M (2025) Agricultural robotics: A technical review addressing challenges in sustainable crop production. Robotics. 14(2):9. Available from: https://www.mdpi.com/2218-6581/14/2/9 Chen S, Xu G, Gao Z (2020) Development of a greenhouse spray robot based on machine vision. Int J Agric Biol Eng 13(1):116–123 Chen S, Xu Z, Zhao D (2021) Design and field evaluation of a sweet pepper harvesting robot. Comput Electron Agric 187:106279 Chlingaryan A, Sukkarieh S, Whelan B (2018) Machine learning approaches for crop yield prediction and nitrogen status estimation in precision agriculture: A review. Comput Electron Agric 151:61–69 Choi S, Kim D, Lee D, Lee J (2022) Review of UAV-based remote sensing in greenhouse crop monitoring. Comput Electron Agric 196:106864 Dario P, Siciliano B, Kragic D (2021) Digital farming: Robotics and AI perspectives. IEEE Rob Autom Magazine 28(3):2–5 Feng Q, Zou J, Li X (2020) Development and performance evaluation of a robotic tomato harvesting system based on deep learning and machine vision. Comput Electron Agric 178:105754 Feng Q, Zhang Q, He L (2022) Automation technologies for intelligent greenhouse crop management. Agric Syst 198:103393 Greenhouse Robots (2022) Ultimate Solutions to Improve Automation in Protected Cropping Systems-A Review. Sustainability 14(11):6436 Guo W, Xue J, Jiang H (2022) Advances in machine vision-based plant disease detection for protected horticulture. Comput Electron Agric 198:107017 Hemming J, Balendonck J (2024) Advances in the use of robotics in greenhouse cultivation. In: Advances in Agri-Food Robotics. Cambridge, UK: Burleigh Dodds Science Publishing; Available from: https://research.wur.nl/en/publications/advances-in-the-use-of-robotics-in-greenhouse-cultivation Hemming J (2020) Current developments in greenhouse robotics and challenges for the future. Acta Hort 1296:975–985. 10.17660/ActaHortic.2020.1296.124 Hemming S, de Zwart F, Elings A, Righini I, Petropoulou A (2019) Remote control of greenhouse vegetable production with artificial intelligence-greenhouse climate, irrigation, and crop production. Sensors 19(8):1807. 10.3390/s19081807 Kamilaris A (2018) Prenafeta-Boldú FX. Deep learning in agriculture: a survey. Comput Electron Agric 147:70–90 Kang H, Lee H, Lee D (2022) Deep-learning-based robotic harvesting of sweet peppers in greenhouse environments. Sensors 22(12):4465 Kang H, Noguchi N, Reid JF (2020) Recent progress in agricultural robotics: A review focusing on harvesting and crop care. Biosyst Eng 196:135–147 Lehnert C, English A, McCool C, Tow AW, Upcroft B (2020) Autonomous sweet pepper harvesting for protected cropping systems. IEEE Robot Autom Lett 5(4):5339–5842 Li Y, Nie L, Wang J (2021) Early detection of tomato diseases using hyperspectral imaging and deep learning. Biosyst Eng 202:135–146 Li Z, Chen Y, Luo X (2021) A review of intelligent harvesting robots for horticultural crops. Comput Electron Agric 187:106259 Li Z, Zhao Y, Zhang C (2020) Design and optimization of robotic systems for greenhouse harvesting: A review. Comput Electron Agric 178:105754 Lim JW, Reza MN, Chung S-O, Lee K-Y, Lee S-Y, Lee B (2023) Application of artificial neural network in smart protected horticulture: A review. Precision Agric Sci Technol 5(1):29–41. 10.12972/pastj.20230003 Meng Q, Xiong J, Zhang Z (2022) Autonomous greenhouse spraying robot: Navigation, control, and performance evaluation. Comput Electron Agric 199:107163 Moher D, Liberati A, Tetzlaff J, Altman DG (2009) PRISMA Group. Preferred Reporting Items for Systematic Reviews and Meta-Analyses: the PRISMA Statement. PLoS Med 6(7):e1000097 Moher D, Liberati A, Tetzlaff J, Altman DG (2009) PRISMA Group. PLoS Med 6(7):e1000097 Moreno JC, Rodríguez F, Sánchez-Hermosilla J, Giménez A, Sánchez-Molina JA (2004) Feasibility analysis of robots in greenhouses: A case study in European Mediterranean countries. Smart Agricultural Technol 9:100638. 10.1016/j.atech.2024.100638 Nawar S (2021) Human-Robot Interaction in Agriculture: A Systematic Review. Comput Electron Agric 186:106151 Pathania J, Verma P, Bodh S, Das S (2022) Role of robotics and artificial intelligence in horticulture for sustainable resource development: A review. Environ Eng Manag J 21(12):2067–2082 Feng Q, Zhang Y, He L (2023) Advances in robotic harvesting systems for greenhouse crops: Technologies and challenges. Comput Electron Agric 204:107547 Qiu R, Sun D, Dong H (2022) Automation and robotics in hydroponic lettuce production: Current status and future perspectives. Horticulturae 8(10):879 Raza SE, Prince G, Bechar A (2023) Advances in robotic sensing and automation for crop monitoring and management. Agric Syst 206:103592 Sa I, Ge Z, Dayoub F, Upcroft B, Perez T, McCool C (2017) DeepFruits: A fruit detection system using deep neural networks. Sensors 17(12):122 Sa I, Popović M, Khanna R, Chen Z, Lottes P, Liebisch F, Nieto J, Siegwart R (2020) WeedMap: A large-scale semantic weed mapping framework using aerial multispectral imaging and deep neural network for precision farming. Remote Sens 10(9):1423 Sánchez-Molina JA, Rodríguez F, Moreno JC, Sánchez-Hermosilla J, Giménez A (2004) Robotics in greenhouses: Scoping review. Comput Electron Agric 219:108750. 10.1016/j.compag.108750 Shamshiri RR, Ahmad D, King AJ, Thorp KR, Hameed IA, Balasundram SK (2021) A review of autonomous agricultural robots and the future of data-driven farming. Agricultural Eng International: CIGR J 23(4):1–20 Shamshiri RR, Kalantari F, Ting KC, Thorp KR, Hameed IA, Weltzien C, Ahmad D (2018) Advances in greenhouse automation and controlled environment agriculture: A transition to plant factories and urban agriculture. Int J Agricultural Biol Eng 11(1):1–22 Shamshiri RR, Weltzien C, Hameed IA (2018) Research and development in agricultural robotics: A perspective of digital farming. Int J Agricultural Biol Eng 11(4):1–14 Shamshiri RR (2018) Advances in greenhouse automation and controlled environment agriculture: A transition to plant factories and smart greenhouses. Int J Agricultural Biol Eng 11(1):1–22 Shamshiri RR, Weltzien C, Hameed IA, Yule IJ, Grift TE, Balasundram SK, Pitonakova L, Ahmad D, Chowdhary G (2018) Research and development in agricultural robotics: A perspective of digital farming. Int J Agricultural Biol Eng 11(4):1–14 Silwal A, Davidson JR, Karkee M, Mo C, Zhang Q (2017) Design, integration, and field evaluation of a robotic apple harvesting system. Robotics 6(4):24 Simpson A, Harvey R, Fox C (2025) A review of adaptable technologies for robotic urban horticulture. Front Sustainable Food Syst 9:1605107. 10.3389/fsufs.2025.1605107 Spagnuolo M, Todde G, Caria M, Furnitto N, Schillaci G, Failla S (2025) Agricultural robotics: A technical review addressing challenges in sustainable crop production. Robotics 14(2):9. 10.3390/robotics14020009 Tagarakis AC, Liakos V, Fountas S (2020) Robotic pruning systems for greenhouse crops: Current status and challenges. Acta Hort 1272:221–228 Tzachor A, Goh J, Tan CS, Chen JK, Soon CF (2023) Agricultural robotics for climate-smart horticulture. npj Clim Atmospheric Sci 6(1):55. 10.1038/s41612-023-00385-7 Tzounis A, Katsoulas N, Bartzanas T, Kittas C (2017) Internet of Things in agriculture, recent advances and future challenges. Biosyst Eng 164:31–48 van Henten EJ, Hemming J, Wouters MWJM, Rukundo M, Edan Y (2023) Robotics and AI for greenhouse horticulture: Challenges and future directions. Trends Food Sci Technol 134:45–59. 10.1016/j.tifs.2023.05.002 van Henten EJ, Hemming J, Bac CW, Edan Y, van Tuijl BA (2019) Robotics in protected cultivation: Challenges and opportunities. Acta Hort 1256:1–98 Xiong J, Jiang H, Zhou J et al (2021) Intelligent spraying systems in protected horticulture: Technologies and applications. Biosyst Eng 206:54–70 Xiong Y, Ge Y, Grimstad L, From PJ (2019) An autonomous strawberry-harvesting robot: design, development, integration, and field evaluation. J Field Robot 36(2):230–247 Zhang X, Wang Y, Liu Z (2025) Key technologies of robotic arms in unmanned greenhouse. Agronomy 15(11):2498 Zhao Y, Hu J, Liu J (2023) Development of a vision-based autonomous pruning robot for greenhouse tomatoes. Comput Electron Agric 209:107825 Zhao Y, Gong L, Huang Y (2021) Advances in robotic harvesting technologies for greenhouse horticultural crops: A review. Comput Electron Agric 185:106134 Additional Declarations The authors declare no competing interests. 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-9716132","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Systematic Review","associatedPublications":[],"authors":[{"id":640394710,"identity":"3a76a79a-4fdf-4722-845d-a926d3ea4b2a","order_by":0,"name":"Semahegn Geremew Abate","email":"data:image/png;base64,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","orcid":"https://orcid.org/0009-0001-5377-4726","institution":"Mekdela Amba University","correspondingAuthor":true,"prefix":"","firstName":"Semahegn","middleName":"Geremew","lastName":"Abate","suffix":""}],"badges":[],"createdAt":"2026-05-14 15:16:01","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-9716132/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9716132/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109334654,"identity":"23567644-3e6c-4cec-9d30-37d508522461","added_by":"auto","created_at":"2026-05-15 16:43:37","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":74075,"visible":true,"origin":"","legend":"\u003cp\u003eStacked bar chart showing publications from 2010 to 2024 across harvesting, monitoring, spraying, pruning, and other tasks.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-9716132/v1/0fd7bbcdd3bffedab411314d.png"},{"id":109334657,"identity":"4064bbb7-891a-42b1-b35e-c08c1850e1c5","added_by":"auto","created_at":"2026-05-15 16:43:37","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":84631,"visible":true,"origin":"","legend":"\u003cp\u003ePie chart showing task distribution: Harvesting (38%), Monitoring (25%), Spraying (18%), Pruning/Training (9%), Other (10%).\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-9716132/v1/742f4c3e1bd1dcdf38f379c9.png"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eIntegration of Robotics and Automation in Protected Horticultural Systems: \u003c/strong\u003eSystematic Review\u003c/p\u003e","fulltext":[{"header":"1. INTRODUCTION","content":"\u003cp\u003eGlobal population growth, rapid urbanization, and increasing demand for high-quality, safe, and sustainably produced food are placing significant pressure on the horticultural sector to enhance productivity and resource-use efficiency. In response, protected horticultural systems such as greenhouses, polytunnels, and vertical farms have emerged as critical components of intensive crop production due to their capacity for year-round cultivation, improved resource management, and resilience to climatic variability (Shamshiri et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Bechar and Vigneault, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). These systems enable precise control of environmental parameters, including temperature, humidity, light, and nutrient supply, thereby supporting consistent production of high-value crops such as tomato, strawberry, cucumber, and lettuce. However, despite these advantages, protected horticulture remains highly labor-intensive, relying heavily on repetitive and time-consuming manual operations, including planting, pruning, monitoring, pest management, and harvesting (Bagagiolo et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; S\u0026aacute;nchez-Molina \u003cem\u003eet al\u003c/em\u003e., 2024).\u003c/p\u003e \u003cp\u003eRecent advances in robotics, automation, and artificial intelligence (AI) are transforming protected horticultural production systems. Robotic technologies offer opportunities for high-precision and repetitive operations such as seeding, transplanting, spraying, pollination, and harvesting, while automation systems integrated with sensors, computer vision, and machine learning enable real-time monitoring and data-driven decision-making (van Henten et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Tzachor et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The convergence of these technologies is facilitating the transition toward intelligent \u0026ldquo;smart greenhouse\u0026rdquo; systems, where cyber-physical integration and Internet of Things (IoT)-based connectivity enhance productivity, sustainability, and operational efficiency (Lim et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSignificant progress has been made in the development of robotic applications for greenhouse environments. For instance, robotic harvesting systems for crops such as tomato and strawberry have demonstrated promising detection and picking performance, although challenges persist for crops with complex canopy structures, such as cucumber and sweet pepper. Additional innovations include autonomous mobile platforms for navigation and logistics, automated pollination systems, and AI-driven climate and irrigation controllers capable of optimizing environmental conditions in real time (Hemming \u003cem\u003eet al\u003c/em\u003e., 2020; Moreno \u003cem\u003eet al\u003c/em\u003e., 2024). Collectively, these technologies have the potential to improve yield, reduce input waste, and minimize dependence on manual labor.\u003c/p\u003e \u003cp\u003eDespite these technological advancements, large-scale commercial adoption remains limited. Technical challenges such as variability in plant structure, occlusion, dynamic lighting conditions, and the need for delicate crop handling continue to affect system reliability. Furthermore, high initial investment costs, uncertain economic returns, and limited standardization constrain widespread deployment. Social considerations, including human-robot interaction, labor displacement, and data governance, also require careful attention.\u003c/p\u003e \u003cp\u003eAlthough existing studies provide valuable technical insights, the literature remains fragmented across specific tasks, crops, and system components. Most previous reviews focus on isolated applications, such as harvesting or navigation, with limited attention to system integration, technology readiness, and sustainability outcomes. This fragmentation restricts a comprehensive understanding of the current state and future potential of robotics and automation in protected horticulture.\u003c/p\u003e \u003cp\u003eTherefore, this systematic review aims to critically synthesize existing research on the integration of robotics and automation technologies in protected horticultural systems. Specifically, it seeks to (i) classify the major robotic and automation technologies employed, (ii) examine the sensing, perception, and control mechanisms underlying their performance, (iii) evaluate their operational efficiency and technology readiness, and (iv) identify key research gaps and future directions. By providing a comprehensive and structured analysis, this review contributes to advancing the development of intelligent, efficient, and sustainable protected horticultural production systems.\u003c/p\u003e"},{"header":"2. METHODS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Protocol and objectives\u003c/h2\u003e \u003cp\u003eThis systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to ensure transparency, reproducibility and methodological rigour (Moher et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). Here are developed a priori research questions to guide the review, for example:\u003c/p\u003e \u003cp\u003eWhat types of robotic and automated technologies have been applied in protected horticultural systems (greenhouses, tunnels, indoor vertical horticulture)?\u003c/p\u003e \u003cp\u003eWhat are the functional tasks: harvesting, pruning, monitoring, navigation) addressed by these systems within protected horticulture?\u003c/p\u003e \u003cp\u003eWhat are reported outcomes (technical performance, economic efficiency, and crop yield/quality, labour saving) and what gaps remain?\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Information sources and search strategy\u003c/h2\u003e \u003cp\u003eWe searched the following electronic bibliographic databases: Scopus, Web of Science (Core Collection), IEEE Xplore, SpringerLink, and Taylor \u0026amp; Francis Online. The search was conducted covering the time period from 1 January 2010 to 31 December 2024. We applied a comprehensive Boolean search string combining synonyms for robotics/automation and protected horticulture.\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\u003eillustrates the databases, number of studies retrieved, and search keywords.\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=\"char\" char=\".\" 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\u003eDatabase\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNumber of Articles Retrieved\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eKeywords/Filters Applied\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eScopus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e450\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003erobot*, automation*, greenhouse, harvesting\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWeb of Science\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e380\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003erobotic system*, controlled environment, monitoring\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIEEE Xplore\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e220\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003erobotic manipulators, sensors, navigation\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSpringerLink\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e150\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eautomation, indoor horticulture\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTaylor \u0026amp; Francis\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e100\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003erobotics, greenhouse, horticultural crops\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\u003eWe supplemented the database search with hand-searching of reference lists of included articles and key review papers. Duplicate records were removed using reference-management software.\u003c/p\u003e \u003cp\u003e \u003cb\u003eEligibility criteria\u003c/b\u003e \u003c/p\u003e \u003cp\u003eWe applied explicit inclusion and exclusion criteria. Included studies met all of the following:\u003c/p\u003e \u003cp\u003ePublished in English in a peer-reviewed journal or full conference proceedings.\u003c/p\u003e \u003cp\u003eFocused on robotic or automated systems (hardware and/or software) implemented within a protected horticultural environment (greenhouse, tunnel, indoor horticulture) rather than open-field only.\u003c/p\u003e \u003cp\u003eReported empirical results (prototype development, pilot trial, technical validation, economic assessment) rather than purely conceptual or conceptual review papers.\u003c/p\u003e \u003cp\u003eExcluded studies comprised: editorials, workshop summaries, theses without peer review, purely open-field systems not specified for protected horticulture, and studies where robotics/automation was only tangentially mentioned without empirical evaluation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Study selection\u003c/h2\u003e \u003cp\u003eAfter removal of duplicates, two independent reviewers screened titles and abstracts for potential relevance. Discrepancies were resolved by discussion or by involving a third reviewer. Full-text screening was then conducted for all articles passing the abstract stage. A PRISMA flow diagram was constructed to document numbers at each stage (identification, screening, eligibility, inclusion).\u003c/p\u003e \u003cp\u003e \u003cb\u003eData extraction and quality assessment\u003c/b\u003e \u003c/p\u003e \u003cp\u003eFrom each included study we extracted the following data: author(s); year of publication; country of study; horticultural crop(s) and protected-environment type; description of robotic/automation system (platform type, sensors, end-effector, navigation strategy, autonomy level); primary agronomic task addressed on harvesting, monitoring, spraying, pruning); stage of development (laboratory prototype, greenhouse pilot, commercial deployment); reported performance metrics (accuracy, cycle time, yield improvement, labour reduction, cost savings); and identified challenges/limitations.\u003c/p\u003e \u003cp\u003eWe assessed the quality of empirical studies using a custom checklist adapted from previous robotics-in-agriculture reviews such as assessment of experimental design, sample size, validation under realistic conditions, reproducibility of results (Nawar \u003cem\u003eet al\u003c/em\u003e., 2021). Studies were classified as high, moderate or low quality based on criteria such as clarity of methods, presence of quantitative evaluation and presence of field or greenhouse trials.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Data synthesis\u003c/h2\u003e \u003cp\u003eData synthesis was descriptive and narrative due to heterogeneity in system types, horticultural crops, evaluation metrics and study designs. Here is the summarised trends by task type, crop type, system maturity, geographic region and reported outcomes. Where sufficient quantitative data were available, we constructed summary tables (autonomy-level vs performance) and visualised frequency counts of tasks and publication year distributions. We also identified gaps and future research opportunities.\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\u003esummarised trends by task type, crop type, system maturity, geographic region and reported outcomes\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\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 \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRobot Type\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAutonomy Level\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCrop\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eKey Outcome\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHarvesting\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMobile manipulator\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSemi-autonomous\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTomato\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e15% yield increase\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMonitoring\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUAV\u0026thinsp;+\u0026thinsp;camera sensors\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAutonomous\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eLettuce\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eEarly disease detection\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSpraying\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGantry-mounted arm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAutonomous\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eStrawberry\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e30% labour reduction\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\u003eSources:(Greenhouse Robots,2022;Advances in Agriculture Robotics,2021).\u003c/p\u003e \u003c/div\u003e"},{"header":"3. RESULTS","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Study Identification, Selection, and PRISMA Flow\u003c/h2\u003e \u003cp\u003eThe database search yielded a total of 1,283 records. After the removal of 247 duplicates, 1,036 studies remained for title and abstract screening. Of these, 816 records were excluded due to lack of relevance, primarily because they focused on open-field agriculture or did not include empirical evaluation of robotic or automated systems in protected horticulture. Consequently, 220 full-text articles were assessed for eligibility, of which 128 studies met all inclusion criteria and were included in the final qualitative synthesis.\u003c/p\u003e \u003cp\u003eThe substantial reduction from initial identification to final inclusion reflects the specificity of robotics applications within protected horticultural systems. The majority of excluded studies addressed general agricultural robotics or unrelated domains, highlighting that research on greenhouse and controlled-environment applications remains relatively specialized. This indicates that, although the field is rapidly evolving, the body of targeted and empirically validated research is still limited.\u003c/p\u003e \u003cp\u003eThe high exclusion rate underscores both the niche nature of robotics in protected horticulture and the need for more focused and application-specific research in this domain.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Temporal, Geographic, and Crop Distribution\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eAnnual publication trend (2010\u0026ndash;2024) by task category\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eYear\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHarvesting\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMonitoring\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eSpraying\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePruning/Training\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eOther\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2010\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2015\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2020\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e2024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e \u003cp\u003e7\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\u003eThe number of publications on robotics and automation in protected horticultural systems has increased markedly since 2015, with a particularly steep rise in studies focused on harvesting tasks. This surge reflects the growing technological feasibility and economic necessity of automated harvesting systems, as global horticulture faces persistent labor shortages and rising labor costs (Bac et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Shamshiri et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Advances in machine vision, soft robotics, and gripper design have made robotic harvesting more practical for delicate crops such as tomatoes, cucumbers, and strawberries (Li et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Zhao et al., \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eMeanwhile, publications related to monitoring and spraying have grown steadily, aligning with the increasing adoption of precision agriculture and environmental monitoring technologies in controlled environments (Sa et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Tzounis et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). These trends suggest a paradigm shift toward data-driven decision-making and automation-enhanced management, particularly in high-value crops grown under greenhouses or polytunnels (van Henten et al., \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe expansion in pruning/training and other robotic tasks after 2020 indicates the diversification of research interests as automation technologies mature. This diversification likely stems from the integration of artificial intelligence (AI) and Internet of Things (IoT) platforms, enabling multifunctional robotic systems capable of performing complex horticultural operations (Chlingaryan et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Bechar \u0026amp; Vigneault, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eOverall, the temporal publication pattern demonstrates a clear acceleration of research activity after 2015, driven by both technological innovation and practical needs within protected horticultural production systems.\u003c/p\u003e \u003cp\u003eHarvesting dominates publications, especially after 2015. Monitoring and spraying also show steady growth, reflecting increased research interest in automated crop management.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Geographic distribution of included studies\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"No\" id=\"Taba\" border=\"1\"\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" 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\u003eRegion\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNumber of studies\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePercentage\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEurope\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e45%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNorth America\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e32\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e25%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAsia\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e26\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e20%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOther regions\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10%\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\u003eEurope is the leading contributor to research on robotics in protected horticulture, likely due to the prevalence of greenhouse production and strong robotics research programs across the region. This is consistent with the finding that about 63% of agricultural robotics manufacturers and research activities are concentrated in Europe, followed by 28% in the Americas, while Asia contributes about 8% and Oceania only 1% (Bongiovanni et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). European countries such as the Netherlands, Germany, and Spain have well-established greenhouse industries supported by major research institutions like Wageningen University \u0026amp; Research, which has specialized programs in vision-based agricultural robotics (Hemming and Balendonck, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eNorth America and Asia also play a significant role, but their contributions remain secondary compared with Europe. Studies have shown that the majority of greenhouse automation projects and prototypes originate from Western Europe, reflecting the region\u0026rsquo;s advanced agricultural technology infrastructure (Hemming and Balendonck, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Meanwhile, regions such as Sub-Saharan Africa and Latin America remain underrepresented in research on robotics and automation for horticultural production, indicating a significant geographic gap (Simpson et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThis imbalance highlights the need for broader international research collaboration and technology adaptation in emerging horticultural regions. Increasing adoption in these regions could address labor challenges and enhance productivity as protected cultivation expands globally.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e3.5 Crop distribution among studies\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCrop distribution among studies\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=\"char\" char=\".\" 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\u003eCrop\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNumber of studies\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ePercentage\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTomato\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e28\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e22%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSweet Pepper\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e19\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e15%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLettuce\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eStrawberry\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOrnamentals/Other\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e42%\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\u003eTomatoes are the most frequently studied crop, likely due to their high economic value, global production volume, and favorable morphology for robotic manipulation, particularly in greenhouse environments. The repetitive growth pattern, well-defined fruit geometry, and dense greenhouse production make tomatoes ideal models for testing robotic harvesting and vision systems (Bac et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Feng et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSweet pepper and lettuce follow, reflecting their high market demand and prevalence in controlled-environment agriculture. Robotic systems for sweet pepper are intensively researched due to the crop\u0026rsquo;s delicate fruit structure and occlusion challenges, which provide valuable test cases for advanced computer vision and end-effector designs (Lehnert et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Chen et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Lettuce, on the other hand, is a major leafy crop in hydroponic and vertical farming systems, where automation enhances uniformity and reduces labor costs (Qiu et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe relatively high proportion of studies on ornamentals and other crops demonstrates a growing interest in non-food applications of robotics, such as transplanting, pruning, and monitoring in nursery and floriculture industries. However, this trend also highlights that many robotic systems remain crop-specific prototypes rather than fully generalizable or commercially deployable solutions (Van Henten et al., \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Shamshiri et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2018\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e3.6 \u003cb\u003eCrop distribution among studies\u003c/b\u003e\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCrop distribution among studies\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\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 \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTask\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRobot Type\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAutonomy Level\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eKey Sensors\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eStage of Development\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eReferences\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHarvesting\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMobile manipulator\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSemi-autonomous\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRGB-D cameras\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003ePilot greenhouse\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eGreenhouse Robots, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2022\u003c/span\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMonitoring\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUAV\u0026thinsp;+\u0026thinsp;camera sensors\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAutonomous\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMultispectral/IR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eLab \u0026amp; Pilot\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eSystematic Review of 59 Field Robots, 2021\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSpraying\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGantry-mounted arm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAutonomous\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRGB\u0026thinsp;+\u0026thinsp;LiDAR\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eGreenhouse pilot\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eDigital Farming robotics perspective, 2021\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePruning\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFixed robotic arm\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSemi-autonomous\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRGB-D\u0026thinsp;+\u0026thinsp;force sensors\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eLab prototype\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eDiscover Sustainability, 2025\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\u003eMost robotic systems in protected horticulture remain at the pilot or prototype stages, with relatively few being deployed commercially. This reflects ongoing technical and economic challenges in achieving robustness under real greenhouse conditions (Bac et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Shamshiri et al., \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Mobile manipulators dominate harvesting applications due to their flexibility and ability to navigate structured environments, while UAVs (Unmanned Aerial Vehicles) are frequently employed for monitoring tasks owing to their rapid, non-invasive data collection capability (Choi et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Dario et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eVision-based sensors, such as RGB-D, multispectral, and infrared cameras, are the most common for perception and decision-making tasks. These sensors are essential for accurate detection, localization, and classification of crops, fruits, and canopy structures (Sa et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Kang et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Autonomy levels vary by task complexity systems performing monitoring and spraying tend to have higher levels of automation due to predefined navigation paths and simpler manipulation requirements, whereas harvesting and pruning tasks still require semi-autonomous control because of the variability in fruit position, occlusion, and plant structure (Arad et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Feng et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e3.7 Reported Outcomes\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 7\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eReported Outcomes\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\u003eTask\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePerformance Gains\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eKey Challenges\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHarvesting\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e58\u0026ndash;80% success rate, yield increase\u0026thinsp;~\u0026thinsp;15%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCrop occlusion, fruit detection, delicate handling\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMonitoring\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDisease detection accuracy\u0026thinsp;\u0026plusmn;\u0026thinsp;10\u0026ndash;15%, early warning\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCost of sensors, data integration\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSpraying\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLabour savings 20\u0026ndash;30%, precise input application\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNavigation in dense canopy, drift control\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePruning\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTrajectory error\u0026thinsp;\u0026lt;\u0026thinsp;0.3 mm, success\u0026thinsp;~\u0026thinsp;93% in simulation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eReal-world unstructured environment, seasonal timing\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\u003eHarvesting and monitoring systems currently exhibit the highest technical maturity in protected horticulture. Robotic harvesters, particularly those employing vision-guided manipulators, report success rates ranging between 58% and 80%, with yield gains of around 15% through reduced crop losses and optimized harvest timing (Bac et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Lehnert et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). However, key challenges remain in fruit occlusion, delicate handling, and the robustness of perception algorithms under varying lighting and crop density conditions (Feng et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eMonitoring systems using multispectral and hyperspectral imaging demonstrate promising disease detection accuracies within \u0026plusmn;\u0026thinsp;10\u0026ndash;15%, offering early-warning capabilities that can prevent severe yield losses (Kamilaris \u0026amp; Prenafeta-Bold\u0026uacute;, 2018; Li et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Despite these advances, the high cost of sensors and challenges in data integration and real-time analytics remain major constraints for commercial deployment (Guo et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFor spraying robots, reported labour savings of 20\u0026ndash;30% and precise input application highlight clear operational benefits (Berenstein \u003cem\u003eet al\u003c/em\u003e., 2020). However, achieving stable navigation in dense canopies and managing drift control under variable microclimatic conditions continue to limit their scalability (Xiong et al., \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Meng et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003ePruning robots exhibit high trajectory accuracy (\u0026lt;\u0026thinsp;0.3 mm) and up to 93% success in simulations (Silwal et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), but transferring these results to real-world, unstructured environments remains difficult due to variability in branch geometry and seasonal timing (Tagarakis et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Zhao et al., \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAcross all task categories, there is a clear research gap in economic assessment, with very few studies addressing cost-effectiveness or return-on-investment (ROI) for robotic systems. Future research should integrate techno-economic evaluations alongside technical metrics to enable broader commercial adoption in protected horticultural systems.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e3.8 Task Distribution and Research Gaps\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab7\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 8\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eTask Distribution and Research Gaps\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=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTasks distribution\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTheir amount in%\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHarvesting\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e38\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMonitoring\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e25\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSpraying\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePruning/Training\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOther tasks\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e10\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\u003eHarvesting tasks dominate the research landscape, accounting for the largest share of robotic studies in protected horticulture. This trend reflects the high labor intensity and economic importance of harvesting operations, which can represent up to 50% of total production costs in horticultural systems (Bac et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Arad et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). The development of robotic harvesters especially for crops like tomato, strawberry, and sweet pepper has thus been prioritized to address challenges related to labor shortages and cost efficiency (Li et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Xiong ., 2019).\u003c/p\u003e \u003cp\u003eMonitoring tasks, including disease and stress detection us\u003cem\u003eet al\u003c/em\u003eing vision and spectral sensors, are the second most common focus, indicating the growing integration of machine vision and data analytics for precision crop management (Raza et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Kamilaris \u0026amp; Prenafeta-Bold\u0026uacute;, 2018). Spraying robots are also emerging, driven by the need for targeted pesticide application and worker safety improvements (Chen et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eHowever, tasks such as pruning, sorting, and post-harvest handling remain underrepresented, despite their importance in quality control and yield optimization (Van Henten et al., \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). The limited research attention to these areas indicates significant opportunities for innovation, especially in developing multi-functional robotic systems capable of performing complex tasks in dynamic greenhouse environments (Lehnert et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFurthermore, the heterogeneity in crop types, robotic platforms, and performance evaluation metrics across studies complicates comparative analyses and hinders the establishment of unified benchmarks. Several authors have emphasized the need for standardized testing environments and performance protocols to enable reproducibility and facilitate meta-analysis (Bechar \u0026amp; Vigneault, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Bac et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Establishing these standards will be crucial for advancing robotics integration and benchmarking progress in protected horticultural systems.\u003c/p\u003e \u003cp\u003e Harvesting is the most studied task, likely due to its labor intensity and economic importance. Other tasks like pruning and post-harvest handling are under explored, representing potential areas for future development.\u003c/p\u003e \u003c/div\u003e"},{"header":"4. CONCLUSION","content":"\u003cp\u003eThis systematic review highlights the rapid advancement of robotics and automation in protected horticultural systems, particularly over the past decade. The analysis of 128 studies indicates that most technological developments are concentrated in harvesting, monitoring, and spraying operations, with harvesting receiving the greatest attention due to its high labor demand and economic significance. These technologies have demonstrated notable improvements in operational efficiency, including increased yield, enhanced precision in crop management, and reduced dependence on manual labor.\u003c/p\u003e \u003cp\u003eHowever, despite promising technical performance, the majority of robotic systems remain at experimental or pilot stages, with limited large-scale commercial adoption. Persistent technical challenges such as crop occlusion, variability in plant structures, and the need for precise and delicate handling continue to constrain system reliability in real greenhouse conditions. In addition, the lack of standardized evaluation frameworks limits comparability across studies and slows technological progress.\u003c/p\u003e \u003cp\u003eEconomic and contextual barriers further restrict adoption, particularly in developing regions. High investment costs, limited infrastructure, and insufficient economic analyses reduce the feasibility of implementing these technologies at scale. The geographic concentration of research in developed regions also reveals a significant gap in innovation and application in emerging horticultural systems.\u003c/p\u003e \u003cp\u003eOverall, while robotics and automation offer substantial potential to enhance productivity, sustainability, and resilience in protected horticulture, future efforts should focus on developing cost-effective, scalable, and adaptable solutions. Greater emphasis on standardization, multi-functional systems, and real-world validation alongside inclusive innovation strategies will be critical to accelerating adoption and ensuring broader global impact.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAdvances in Agriculture Robotics (2021) A State-of-the-Art Review and Challenges Ahead. Robotics 10(2):52\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eApplication of Industry-Technologies in Horticulture (2022) Horticulture-Adoption of Industry-Technologies in Horticulture for Meeting Sustainable Farming. Appl Sci 12(24):12557. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/app122412557\u003c/span\u003e\u003cspan address=\"10.3390/app122412557\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArad B, Ben-Shahar O, Stern H, Edan Y (2023) Performance evaluation of robotic harvesters in structured greenhouse environments. Comput Electron Agric 206:107668\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArad B, Balendonck J, Barth R, Ben-Shahar O, Edan Y, van Henten EJ (2020) Development of a sweet pepper harvesting robot. J Field Robot 37(6):1027\u0026ndash;1043\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBac CW, van Henten EJ, Edan Y (2020) Key performance indicators for robotic harvesting of greenhouse crops. Biosyst Eng 191:60\u0026ndash;75\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBac CW, Hemming J, van Henten EJ (2017) Harvesting robots for high-value crops: State-of-the-art review and challenges ahead. Biosyst Eng 153:96\u0026ndash;111\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBac CW, van Henten EJ, Hemming J, Edan Y (2017) Harvesting robots for high-value crops: State-of-the-art review and challenges ahead. J Field Robot 34(6):1199\u0026ndash;1235\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBagagiolo G, Matranga G, Cavallo E, Pampuro N (2022) Greenhouse Robots: Ultimate Solutions to Improve Automation in Protected Cropping Systems-A Review. Sustainability 14(11):6436. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/su14116436\u003c/span\u003e\u003cspan address=\"10.3390/su14116436\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBechar A, Vigneault C (2017) Agricultural robots for field operations. Part 2: Operations and systems. Biosyst Eng 153:110\u0026ndash;128. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.biosystemseng.2016.11.004\u003c/span\u003e\u003cspan address=\"10.1016/j.biosystemseng.2016.11.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBerenstein R, Edan Y (2020) Automation in protected agriculture: The case of robotic spraying in greenhouses. Biosyst Eng 197:135\u0026ndash;152\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoccia G, Pasquini G, Moretti G (2022) Greenhouse robots: ultimate solutions to improve automation in protected cropping systems-A review. Sustainability 14(11):6436. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/su14116436\u003c/span\u003e\u003cspan address=\"10.3390/su14116436\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBongiovanni R, Di Gennaro S, Perez-Ruiz M (2025) Agricultural robotics: A technical review addressing challenges in sustainable crop production. \u003cem\u003eRobotics.\u003c/em\u003e 14(2):9. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.mdpi.com/2218-6581/14/2/9\u003c/span\u003e\u003cspan address=\"https://www.mdpi.com/2218-6581/14/2/9\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen S, Xu G, Gao Z (2020) Development of a greenhouse spray robot based on machine vision. Int J Agric Biol Eng 13(1):116\u0026ndash;123\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen S, Xu Z, Zhao D (2021) Design and field evaluation of a sweet pepper harvesting robot. Comput Electron Agric 187:106279\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChlingaryan A, Sukkarieh S, Whelan B (2018) Machine learning approaches for crop yield prediction and nitrogen status estimation in precision agriculture: A review. Comput Electron Agric 151:61\u0026ndash;69\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChoi S, Kim D, Lee D, Lee J (2022) Review of UAV-based remote sensing in greenhouse crop monitoring. Comput Electron Agric 196:106864\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDario P, Siciliano B, Kragic D (2021) Digital farming: Robotics and AI perspectives. IEEE Rob Autom Magazine 28(3):2\u0026ndash;5\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFeng Q, Zou J, Li X (2020) Development and performance evaluation of a robotic tomato harvesting system based on deep learning and machine vision. Comput Electron Agric 178:105754\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFeng Q, Zhang Q, He L (2022) Automation technologies for intelligent greenhouse crop management. Agric Syst 198:103393\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGreenhouse Robots (2022) Ultimate Solutions to Improve Automation in Protected Cropping Systems-A Review. Sustainability 14(11):6436\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuo W, Xue J, Jiang H (2022) Advances in machine vision-based plant disease detection for protected horticulture. Comput Electron Agric 198:107017\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHemming J, Balendonck J (2024) Advances in the use of robotics in greenhouse cultivation. In: \u003cem\u003eAdvances in Agri-Food Robotics.\u003c/em\u003e Cambridge, UK: Burleigh Dodds Science Publishing; Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://research.wur.nl/en/publications/advances-in-the-use-of-robotics-in-greenhouse-cultivation\u003c/span\u003e\u003cspan address=\"https://research.wur.nl/en/publications/advances-in-the-use-of-robotics-in-greenhouse-cultivation\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHemming J (2020) Current developments in greenhouse robotics and challenges for the future. Acta Hort 1296:975\u0026ndash;985. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.17660/ActaHortic.2020.1296.124\u003c/span\u003e\u003cspan address=\"10.17660/ActaHortic.2020.1296.124\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHemming S, de Zwart F, Elings A, Righini I, Petropoulou A (2019) Remote control of greenhouse vegetable production with artificial intelligence-greenhouse climate, irrigation, and crop production. Sensors 19(8):1807. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/s19081807\u003c/span\u003e\u003cspan address=\"10.3390/s19081807\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKamilaris A (2018) Prenafeta-Bold\u0026uacute; FX. Deep learning in agriculture: a survey. Comput Electron Agric 147:70\u0026ndash;90\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKang H, Lee H, Lee D (2022) Deep-learning-based robotic harvesting of sweet peppers in greenhouse environments. Sensors 22(12):4465\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKang H, Noguchi N, Reid JF (2020) Recent progress in agricultural robotics: A review focusing on harvesting and crop care. Biosyst Eng 196:135\u0026ndash;147\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLehnert C, English A, McCool C, Tow AW, Upcroft B (2020) Autonomous sweet pepper harvesting for protected cropping systems. IEEE Robot Autom Lett 5(4):5339\u0026ndash;5842\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Y, Nie L, Wang J (2021) Early detection of tomato diseases using hyperspectral imaging and deep learning. Biosyst Eng 202:135\u0026ndash;146\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Z, Chen Y, Luo X (2021) A review of intelligent harvesting robots for horticultural crops. Comput Electron Agric 187:106259\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Z, Zhao Y, Zhang C (2020) Design and optimization of robotic systems for greenhouse harvesting: A review. Comput Electron Agric 178:105754\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLim JW, Reza MN, Chung S-O, Lee K-Y, Lee S-Y, Lee B (2023) Application of artificial neural network in smart protected horticulture: A review. Precision Agric Sci Technol 5(1):29\u0026ndash;41. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.12972/pastj.20230003\u003c/span\u003e\u003cspan address=\"10.12972/pastj.20230003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeng Q, Xiong J, Zhang Z (2022) Autonomous greenhouse spraying robot: Navigation, control, and performance evaluation. Comput Electron Agric 199:107163\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoher D, Liberati A, Tetzlaff J, Altman DG (2009) PRISMA Group. Preferred Reporting Items for Systematic Reviews and Meta-Analyses: the PRISMA Statement. PLoS Med 6(7):e1000097\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoher D, Liberati A, Tetzlaff J, Altman DG (2009) PRISMA Group. PLoS Med 6(7):e1000097\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoreno JC, Rodr\u0026iacute;guez F, S\u0026aacute;nchez-Hermosilla J, Gim\u0026eacute;nez A, S\u0026aacute;nchez-Molina JA (2004) Feasibility analysis of robots in greenhouses: A case study in European Mediterranean countries. Smart Agricultural Technol 9:100638. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.atech.2024.100638\u003c/span\u003e\u003cspan address=\"10.1016/j.atech.2024.100638\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNawar S (2021) Human-Robot Interaction in Agriculture: A Systematic Review. Comput Electron Agric 186:106151\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePathania J, Verma P, Bodh S, Das S (2022) Role of robotics and artificial intelligence in horticulture for sustainable resource development: A review. Environ Eng Manag J 21(12):2067\u0026ndash;2082\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFeng Q, Zhang Y, He L (2023) Advances in robotic harvesting systems for greenhouse crops: Technologies and challenges. Comput Electron Agric 204:107547\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQiu R, Sun D, Dong H (2022) Automation and robotics in hydroponic lettuce production: Current status and future perspectives. Horticulturae 8(10):879\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRaza SE, Prince G, Bechar A (2023) Advances in robotic sensing and automation for crop monitoring and management. Agric Syst 206:103592\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSa I, Ge Z, Dayoub F, Upcroft B, Perez T, McCool C (2017) DeepFruits: A fruit detection system using deep neural networks. Sensors 17(12):122\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSa I, Popović M, Khanna R, Chen Z, Lottes P, Liebisch F, Nieto J, Siegwart R (2020) WeedMap: A large-scale semantic weed mapping framework using aerial multispectral imaging and deep neural network for precision farming. Remote Sens 10(9):1423\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eS\u0026aacute;nchez-Molina JA, Rodr\u0026iacute;guez F, Moreno JC, S\u0026aacute;nchez-Hermosilla J, Gim\u0026eacute;nez A (2004) Robotics in greenhouses: Scoping review. Comput Electron Agric 219:108750. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.compag.108750\u003c/span\u003e\u003cspan address=\"10.1016/j.compag.108750\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShamshiri RR, Ahmad D, King AJ, Thorp KR, Hameed IA, Balasundram SK (2021) A review of autonomous agricultural robots and the future of data-driven farming. Agricultural Eng International: CIGR J 23(4):1\u0026ndash;20\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShamshiri RR, Kalantari F, Ting KC, Thorp KR, Hameed IA, Weltzien C, Ahmad D (2018) Advances in greenhouse automation and controlled environment agriculture: A transition to plant factories and urban agriculture. Int J Agricultural Biol Eng 11(1):1\u0026ndash;22\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShamshiri RR, Weltzien C, Hameed IA (2018) Research and development in agricultural robotics: A perspective of digital farming. Int J Agricultural Biol Eng 11(4):1\u0026ndash;14\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShamshiri RR (2018) Advances in greenhouse automation and controlled environment agriculture: A transition to plant factories and smart greenhouses. Int J Agricultural Biol Eng 11(1):1\u0026ndash;22\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShamshiri RR, Weltzien C, Hameed IA, Yule IJ, Grift TE, Balasundram SK, Pitonakova L, Ahmad D, Chowdhary G (2018) Research and development in agricultural robotics: A perspective of digital farming. Int J Agricultural Biol Eng 11(4):1\u0026ndash;14\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSilwal A, Davidson JR, Karkee M, Mo C, Zhang Q (2017) Design, integration, and field evaluation of a robotic apple harvesting system. Robotics 6(4):24\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSimpson A, Harvey R, Fox C (2025) A review of adaptable technologies for robotic urban horticulture. Front Sustainable Food Syst 9:1605107. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fsufs.2025.1605107\u003c/span\u003e\u003cspan address=\"10.3389/fsufs.2025.1605107\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSpagnuolo M, Todde G, Caria M, Furnitto N, Schillaci G, Failla S (2025) Agricultural robotics: A technical review addressing challenges in sustainable crop production. Robotics 14(2):9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/robotics14020009\u003c/span\u003e\u003cspan address=\"10.3390/robotics14020009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTagarakis AC, Liakos V, Fountas S (2020) Robotic pruning systems for greenhouse crops: Current status and challenges. Acta Hort 1272:221\u0026ndash;228\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTzachor A, Goh J, Tan CS, Chen JK, Soon CF (2023) Agricultural robotics for climate-smart horticulture. npj Clim Atmospheric Sci 6(1):55. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41612-023-00385-7\u003c/span\u003e\u003cspan address=\"10.1038/s41612-023-00385-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTzounis A, Katsoulas N, Bartzanas T, Kittas C (2017) Internet of Things in agriculture, recent advances and future challenges. Biosyst Eng 164:31\u0026ndash;48\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Henten EJ, Hemming J, Wouters MWJM, Rukundo M, Edan Y (2023) Robotics and AI for greenhouse horticulture: Challenges and future directions. Trends Food Sci Technol 134:45\u0026ndash;59. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.tifs.2023.05.002\u003c/span\u003e\u003cspan address=\"10.1016/j.tifs.2023.05.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Henten EJ, Hemming J, Bac CW, Edan Y, van Tuijl BA (2019) Robotics in protected cultivation: Challenges and opportunities. Acta Hort 1256:1\u0026ndash;98\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXiong J, Jiang H, Zhou J et al (2021) Intelligent spraying systems in protected horticulture: Technologies and applications. Biosyst Eng 206:54\u0026ndash;70\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXiong Y, Ge Y, Grimstad L, From PJ (2019) An autonomous strawberry-harvesting robot: design, development, integration, and field evaluation. J Field Robot 36(2):230\u0026ndash;247\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang X, Wang Y, Liu Z (2025) Key technologies of robotic arms in unmanned greenhouse. Agronomy 15(11):2498\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao Y, Hu J, Liu J (2023) Development of a vision-based autonomous pruning robot for greenhouse tomatoes. Comput Electron Agric 209:107825\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao Y, Gong L, Huang Y (2021) Advances in robotic harvesting technologies for greenhouse horticultural crops: A review. Comput Electron Agric 185:106134\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Mekdela Amba University","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":false,"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":"Horticulture, Precision Agriculture, Automation, Sustainable, Crop Management","lastPublishedDoi":"10.21203/rs.3.rs-9716132/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9716132/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe integration of artificial intelligence (AI) into horticultural production is transforming the way crops are monitored, managed, and optimized for productivity and sustainability. This systematic review synthesizes recent developments (2018\u0026ndash;2025) in AI-driven horticultural systems, focusing on machine learning, deep learning, computer vision, and robotics. The findings reveal that AI technologies have significantly advanced in precision phenotyping, disease and pest detection, irrigation and nutrient management, robotic harvesting, and supply-chain optimization. These innovations contribute to enhanced resource efficiency, reduced labor dependence, and improved decision-making accuracy in both open-field and protected cultivation systems. However, challenges persist, including limited access to high-quality datasets, poor model generalization across environments, high implementation costs, and the need for explainable and trustworthy AI systems. Future progress depends on developing open, standardized datasets, scalable low-cost sensor-AI integration for smallholders, and interdisciplinary frameworks that ensure equitable technology adoption. Overall, AI holds transformative potential to make horticultural production more productive, resilient, and sustainable-advancing the global shift toward data-driven and climate-smart agriculture.\u003c/p\u003e","manuscriptTitle":"Integration of Robotics and Automation in Protected Horticultural Systems: Systematic Review","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-15 16:43:34","doi":"10.21203/rs.3.rs-9716132/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":"f955633a-d333-4211-af7d-f3803a7b5a16","owner":[],"postedDate":"May 15th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":68148813,"name":"Horticulture"}],"tags":[],"updatedAt":"2026-05-15T16:43:34+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-15 16:43:34","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9716132","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9716132","identity":"rs-9716132","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}