Impact of Hip Exoskeleton Assistance on Human Gait: A Kinematic Study

Impact of Hip Exoskeleton Assistance on Human Gait: A Kinematic Study | 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 Research Article Impact of Hip Exoskeleton Assistance on Human Gait: A Kinematic Study Nicol Moscatelli, Valentina Lanzani, Cristina Brambilla, Lorenzo Molinari Tosatti, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8367199/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 Wearable robotic devices such as lower-limb exoskeletons have been recently proposed to support mobility, assist activities of daily living, and even aid rehabilitation. Understanding and quantifying the quality of interaction between human and robot is essential to ensure physiological and effective support. In this study, we evaluated a transparent (i.e., non-assisting) modality and two levels of assistance (low and high) of the Hypershell-X hip exoskeleton during overground walking in a cohort of 16 healthy individuals. Our goal was to characterize the impact of the assistance on physiological gait from a kinematic perspective. The results suggest that the exoskeleton preserved natural gait characteristics across conditions: no significant changes were found in spatiotemporal parameters, and joint profiles on the sagittal plane remained highly correlated (r > 0.9). Significant alterations emerged at the highest assistance level, which mainly affected the hip joint, where the assistive torque is directly applied. However, even in this experimental condition, the overall gait biomechanics was preserved. These findings indicate that the Hypershell-X provides functional and effective assistance without relevant alteration of the physiological structure of gait. Exoskeletons Human-Robot Interaction Gait Assistive Devices Wearable Robotics Lower-limb kinematics Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 1. Introduction Wearable lower-limb exoskeletons have emerged as a versatile technology for a wide range of applications. In clinical settings, they are widely used to promote and support motor recovery in individuals with neurological impairments (Carpinella et al., 2009 ; Fritz et al., 2019 ; Gassert & Dietz, 2018 ; Nepomuceno et al., 2024 ; Park et al., 2024 ), and many studies have demonstrated that exoskeleton-based rehabilitation can be comparable, or even superior, to conventional physiotherapy treatments (Calabrò et al., 2021 ; Calafiore et al., 2022 ; Hsu et al., 2023 ). Moreover, lower-limb exoskeletons have been used for gait training and daily assistance in individuals with above-knee amputations (Ishmael et al., 2023 ; Sanz-Morère et al., 2021 ), who often exhibit abnormal gait patterns and compensatory movement strategies. Beyond rehabilitation, exoskeletons are increasingly adopted in non-clinical contexts. Industrial wearable devices, for example, can support workers in physically demanding tasks, helping to reduce fatigue and the risk of developing musculoskeletal disorders (Bogue, 2015 ; Cardoso et al., 2024 ; de Looze et al., 2016 ). More recently, interest has also grown in sports and recreational applications, where wearable assistive technologies aim to reduce energy expenditure (Grimmer & Zhao, 2024 ; Sawicki et al., 2020 ), improve performance (Yandell et al., 2019 ), and prevent injuries (Nurse et al., 2025 ). Lower-limb exoskeletons typically provide assistance at the hip, at the ankle, or at multiple joints (Bryan et al., 2021 ). Among these configurations, hip exoskeletons have shown consistent reductions in the metabolic cost of walking in healthy individuals compared to unassisted gait (Ding et al., 2018 ; Lee et al., 2017 ; Lim et al., 2019 ; Seo et al., 2016 ), and these reductions are often larger than those reported with ankle exoskeletons (Sawicki et al., 2020 ). A key factor contributing to this aspect is the location of the device’s added mass: hip exoskeletons position most of their components close to the trunk and near the body’s center of mass, which helps minimize negative effects and disturbances to natural gait biomechanics (Chen et al., 2018 ; Ishmael et al., 2023 ; Sawicki et al., 2020 ). For these reasons, hip exoskeletons are emerging as promising solutions for improving walking efficiency while preserving comfort and natural movement. Within this context, the Hypershell-X exoskeleton (Hypershell, Shanghai, China) is a lightweight (2.0 Kg) hip exoskeleton designed for assisting a large variety of recreational tasks, including gait, cycling, stairs ascent and descent, mountain climbing, running, and fast walking, with the goal of reducing physical effort by 30%. As these technologies continue to evolve, understanding how exoskeleton assistance influences human movement has become essential for optimizing human-robot interaction. In this regard, kinematic analysis plays a crucial role: while metrics such as metabolic cost are valuable for assessing the effectiveness of the assistance, a complete kinematic analysis provides deeper insights into the quality of the interaction with the device. Spatiotemporal parameters such as cadence and walking speed, combined with joint angle trajectories and coordination patterns, reflect whether the exoskeleton supports the user without compromising the physiological characteristics of gait. In this context, the present pilot study evaluates the interaction between the Hypershell-X exoskeleton and healthy adults during overground walking. By performing a detailed kinematic analysis, we aim to characterize how different levels of robotic assistance modulate human movement, coordination, and functional performance. 2. Materials and Methods 2.1. Equipment The Hypershell-X exoskeleton is an AI-powered wearable device designed to augment human mobility and reduce fatigue during recreational activities. Its control system detects movement patterns in real time and provides adaptive power assistance. The device integrates a motor with 800W peak power and 32Nm torque, delivering substantial support tailored to the specific movement being performed. The Hypershell X includes several assistance modes, each optimized for different activities. The “Eco” mode is the default setting and is programmed to support moderate activities such as walking or hiking. The “Hyper” mode, instead, offers more dynamic assistance for higher intensity tasks, such as running or climbing. Moreover, a “Transparent” mode is also available, allowing the user to temporarily suspend assistance. Both the “Eco” and “Hyper” modes offer four adjustable assistance levels, enabling the user to tailor the amount of support to the task intensity. 2.2. Participants A total of 16 healthy participants (9 males, mean age 37.6 ± 13 years, mean height 180 ± 6.5 cm, mean weight 74.4 ± 6.8 kg; 7 females, mean age 26.1 ± 1.7 years, mean height 165 ± 5.3 cm, mean weight 51.3 ± 3.4 kg) were recruited for this study. None of the participants reported any lower-limb injuries, surgical interventions, neurological or musculoskeletal disorders that could affect their gait. All subjects except two were right-limb dominant, which was determined by their preferred foot used to initiate walking. Before participation, all individuals provided written informed consent in accordance with protocols approved by the Ethics Committee of the National Research Council (approval number N°231878/2025, approved on 24/06/2025) and in compliance with the Declaration of Helsinki. 2.3. Experimental procedure and data recording All participants performed level-ground walking along a 10-meter straight path at a self-selected speed while maintaining a level head posture, as during normal walking. Subjects started from a designated point on the floor and walked to a marker 10 meters away, then stopped, turned around, and repeated the task, completing a total of 6 trials. Walking sessions were repeated under four experimental conditions: a. Free (F): walking without the exoskeleton b. Transparent (T): walking while wearing Hypershell-X, with no active assistance c. Assisted : walking with active assistance provided by the exoskeleton. Two different assistance levels were tested, low (level 1 – A1) and high (level 3 – A3), for the “Eco” mode of the Hypershell-X. The A3 level was intentionally included, even though it was suggested for more demanding activities, with the aim of evaluating how physiological gait is affected when providing a high level of assistance. Data acquisition was carried out in the corridor located in front of the Bio-SimPro-Lab (Advanced Methods for Biomedical Signal and Image Processing Laboratory) at the National Research Council in Milan, Italy. Kinematic data were collected using seven inertial measurement units (IMUs) provided by Captiks S.r.l. (Rome, Italy). One IMU was placed around the waist, and three sensors were positioned on each leg at the foot, ankle and thigh (Fig. 1 ). This configuration enabled the reconstruction of the three degrees of freedom of the pelvis and of the hip, knee and ankle joints. Electromyographic (EMG) signals were also recorded during the trials; however, EMG data were not included in the analysis presented in this study. 2.4. Data processing and outcome metrics Joint angles obtained from the IMUs were pre-processed using a 5th -order, low-pass Butterworth filter, with a cutting frequency of 10 Hz. The segmentation of the recordings into individual gait cycles was performed using the Captiks Motion Analyzer software, which also allowed the extraction of the spatiotemporal gait parameters. Specifically, the following parameters were considered: cadence (steps/min); walking speed (m/s); normalized walking speed (%/s), obtained by normalizing walking speed by the participant’s height; step length (m), defined as the distance between the heel strike of one foot and the subsequent heel strike of the opposite foot; stride length (m), defined as the distance between two consecutive heel strikes of the same foot; stance time (s), defined as the time interval between heel strike and toe-off; swing time (s), defined as the time interval between the toe-off and the next heel strike; stride time (s), defined as the duration of the full gait cycle. Since only healthy individuals were recruited, gait symmetry between limbs was assumed, and therefore we performed the analysis using data from the dominant limb only. In addition to the spatiotemporal parameters, hip, knee, and ankle flexion-extension angles were used to compute the Range of Motion (RoM) for each joint, defined as the difference between the maximum and minimum angles within the gait cycle. To further assess potential changes in gait patterns across different assistance modes, skewness was computed from the joint angle waveforms. Skewness quantifies the temporal asymmetry of a signal around its mean (Srivatsa et al., 2025 ). Negative values indicate that the signal remains predominantly below its mean, which, for joint angles, may reflect a greater proportion of time spent in a joint extension phase. Conversely, positive values indicate that the signal remains predominantly above the mean, suggesting more time spent in flexion. Values close to zero reflect a more balanced, symmetric distribution around the mean (Fraiwan & Hassanin, 2021 ; Marimon et al., 2024 ). Skewness was computed using the skewness built-in function in MATLAB (MathWorks, Natick, USA). All variables were first tested for normality using the Shapiro-Wilk test. Since all variables met the normality assumption, a repeated-measures ANOVA was performed, with the assistance condition (Free Movement (F), Transparent (T), Assistance Level 1 (A1), Assistance Level 3 (A3)) considered as the main factor. In addition, Spearman’s correlation coefficient was computed between joint angle waveforms across conditions to assess the similarity and coherence of gait patterns. Statistical significance was set at p < 0.05. 3. Results Figure 2 illustrates the joint angles on the sagittal plane for a representative subject. Although the mean pattern was consistent across all participants, we chose to present the data from a single subject to provide a cleaner and more readable visualization. This choice also makes it easier to highlight relevant gait events, such as the toe-off instants across conditions. Joint angle trajectories for a representative participant, shown as mean ± standard deviation of the gait cycle phase. The Free condition was used as the reference to compare the other three modalities. In each plot, solid, coloured lines indicate the mean joint angle, while the shaded areas represent the corresponding standard deviations. Coloured, dashed lines mark the toe-off event, separating the stance from the swing phase, and each line matches the colour of its corresponding condition. Hip FE = hip flexion-extension, Knee FE = knee flexion-extension, Ankle DP = ankle dorsiflexion-plantarflexion. F = Free , in green; T = Transparent , in orange; A1 = Assisted Level 1 , in purple; A3 = Assisted Level 3 , in pink. The correlation between the sagittal angle profiles of the main joints (hip, knee, and ankle) across the four different walking conditions (F, T, A1, and A3) was very high (Fig. 3 ). The correlation values between conditions, averaged across all subjects, were all above 0.9 (p < 0.001), indicating a strong similarity and preservation of joint movement patterns despite the different levels of assistance provided by the exoskeleton. The mean values for the spatiotemporal parameters across the four conditions are reported in Table 1 . No significant differences were found, indicating that the overall gait pattern was not altered by the different levels of assistance provided. Table 1 Spatiotemporal parameters across the four walking conditions (averaged between subjects). Each value is expressed as mean ± standard deviation across all participants. Free Transparent Assisted 1 Assisted 3 Cadence (steps/min) 99.2 ± 8.35 99.6 ± 9.66 99.8 ± 9.86 101.6 ± 12.11 Walking speed (m/s) 1.01 ± 0.14 0.96 ± 0.15 1.02 ± 0.16 1.03 ± 0.21 Normalized walking speed (%/s) 58.4 ± 7.48 55.4 ± 7.54 58.9 ± 9.58 59.5 ± 12.65 Step length (m) 0.60 ± 0.08 0.56 ± 0.09 0.59 ± 0.08 0.60 ± 0.09 Stride length (m) 1.23 ± 0.14 1.16 ± 0.15 1.22 ± 0.14 1.21 ± 0.16 Stance time (s) 0.75 ± 0.08 0.74 ± 0.09 0.74 ± 0.10 0.72 ± 0.10 Swing time (s) 0.48 ± 0.04 0.48 ± 0.05 0.47 ± 0.04 0.48 ± 0.05 Stride time (s) 1.22 ± 0.11 1.22 ± 0.12 1.22 ± 0.13 1.20 ± 0.14 Repeated-measures ANOVA revealed significant changes in hip (p = 0.003) and knee (p < 0.001) flexion-extension RoM (Fig. 4 ). For the hip, post-hoc comparisons showed larger RoM at the highest assistance level: significance was met between F and A3 (p = 0.024), T and A3 (p = 0.024), and between the two assistance levels (p = 0.016). Similarly, for the knee, higher RoM was registered in A3 modality compared to the other conditions (p = 0.016 for F, p = 0.002 for T, p = 0.014 for A1). Additionally, a significant difference was also found between the T and the A1 modalities, with the latter reaching higher values. Ankle RoM did not reveal significant changes, suggesting that the interaction with the exoskeleton did not alter the angular trajectories of the distal joints. Skewness resulted significantly altered only for the hip flexion-extension angle (p < 0.001). Specifically, no significant changes were observed between F and T modalities. However, when assistance was provided, both at A1 (p = 0.019 for F, p = 0.026 for T) and at A3 (p = 0.003 for both F and T), skewness significantly increased (Fig. 5 ). This shift brought the values closer to zero, indicating that the joint curve became more symmetrical compared to the baseline condition, where the distribution was negatively skewed. 4. Discussion 4.1. Summary of the main results The goal of this study was to evaluate the interaction between the Hypershell-X hip exoskeleton and a group of healthy participants during overground walking, with a focus on potential deviations from physiological gait. To this end, 16 participants performed four walking sessions under different levels of exoskeleton assistance, and joint kinematics were recorded and analysed. Our results showed no significant changes in spatiotemporal parameters and high correlation in the joint angular patterns across conditions, even if the highest assistance level led to significant alterations in hip and knee RoM and in hip skewness. 4.2. Effects of minimal assistance on gait kinematics When comparing F and T conditions, no significant differences emerged in any of the spatiotemporal or kinematic parameters analysed, suggesting that the device did not interfere with physiological gait when no assistance was provided. This is an encouraging result, as it indicates that the overall structure and weight of the device do not hinder natural walking, which is an essential prerequisite for ensuring that any assistance delivered can be effectively integrated into the user’s movement (Camardella et al., 2021 ; Moscatelli et al., 2025 ; Proietti et al., 2016 ; Stramel & Agrawal, 2022 ). Therefore, we can conclude that the device’s transparency is high when only wearing Hypershell. Even when low-level assistance (A1) was introduced, both spatiotemporal parameters and joint RoMs remained comparable to the free condition, supporting the hypothesis that the A1 modality does not alter physiological gait or the natural amplitude of joint motion. This aligns with the intended purpose of A1, which is designed for daily-life walking and therefore aims to provide minimal, subtle assistance. In addition, the angular profiles remained highly correlated across conditions (r > 0.9), confirming that the overall shape and timing of the joint trajectories were mostly preserved. Interestingly, skewness revealed a subtler modification of hip kinematics. In physiological gait, the hip flexion-extension angle naturally exhibits a negative skewness, reflecting that the joint spends a longer portion of the gait cycle in an extended configuration rather than in a flexed one. This pattern was maintained when assistance was not provided in the Transparent condition. In the A1 assistance configuration, however, skewness became significantly less negative, indicating a shift toward a more symmetric temporal distribution. This suggests that participants spent more time in a flexed configuration compared to the free condition. Such behaviour is consistent with the assistance strategy of the device: although the RoM did not change, the exoskeleton likely supported hip flexion enough to “shift” the angle profile upward (i.e., the hip was maintained in slightly greater flexion without altering the overall curve), thus increasing the time spent in flexion and making the distribution more symmetric. In summary, low-level assistance (A1) preserves the global gait pattern, as reflected by the unchanged RoM, stable spatiotemporal parameters, and high correlation of joint trajectories. This is consistent with previous findings in the literature: in fact, several studies reported no significant alterations in gait kinematics across different assistance levels provided with other devices for gait assistance, even in more complex scenarios such as when using the Lokomat, a lower-limb exoskeleton commonly employed for rehabilitation (Cherni et al., 2023 ; Di Tommaso et al., 2023 ). In our case, the only detectable adaptation was a redistribution of the hip motion phases, possibly reflecting a mild adjustment to the assistance profile without affecting overall kinematic structure. The knee and ankle joints, instead, were not influenced by this assistance level, consistent with the fact that the exoskeleton directly acts on the hip, and further supporting the minimal impact of A1 on physiological gait. 4.3. Effects of high assistance on gait kinematics Unlike what we observed for the A1 assistance level, which was recommended for low-intensity activities such as ground-level walking, significant alterations emerged when the exoskeleton provided a higher amount of assistance. The most evident change was the increase in hip RoM, driven by excessive flexion during the swing phase induced by the stronger flexion torque delivered by the device. Skewness was also affected, showing a shift toward less negative values, and therefore a more symmetric distribution of joint activation. In this case, the effect was much more pronounced than in the A1 condition: the higher assistance level generated a greater flexion peak during swing, combined with a reduced extension in late stance, resulting in an overall increase in the time spent in flexion. Interestingly, the knee was also influenced by the high assistance level. The increased hip flexion lifted the limb more prominently during swing, leading to a significant increase in knee RoM compared to all other conditions. However, this modification was limited to the RoM, as knee skewness remained unchanged: knee flexion-extension angle is naturally characterized by a positive skewness, and the increased RoM at the knee did not substantially modify the temporal distribution of the curve. The ankle joint, instead, did not exhibit significant alterations, confirming that the effects of excessive assistance remain localized to the proximal joints, where the torque is directly applied. Despite these joint-specific changes, neither the overall joint angular shape nor the spatiotemporal parameters were altered. The fundamental characteristics of gait were therefore preserved, even in high assistance trials. This suggests that when the device delivers an excessive amount of torque, the global gait pattern remains stable, while intrinsic features of the joints, particularly at the hip, where the exoskeleton acts directly, are modified. As a result, gait becomes less natural, although from a biomechanical perspective, the system still operates effectively. This observation also suggests that such a high assistance level might be more suitable for more demanding tasks, such as running or climbing, where a stronger flexion contribution is expected. In this scenario, no major kinematic alterations are expected, but rather a form of assistance that is coherent with the increased mechanical and energetic requirements of the task. 4.4. Limitations and future works This study presented a complete kinematic analysis of human gait during different levels of interaction with a powered hip exoskeleton. Although the results obtained indicate that the interaction with the exoskeleton led to minimal kinematic changes, this work did not directly investigate the underlying neural and motor control adaptation potentially induced by the exoskeleton. In particular, valuable information could be gained by analysing individual muscle activation profiles, which are commonly employed to study human-robot interaction (Moscatelli et al., 2025 ), as well as muscle coordination patterns, i.e., muscle synergies, obtained from EMG data (d’Avella et al., 2006 ; Ivanenko et al., 2004 ). Such an approach would allow a deeper understanding of how the central nervous system adapts to the assistance and could provide stronger evidence supporting our results. Another limitation is that the study examined only a single, low-demanding walking task performed at a self-selected speed. As a result, the behaviour of the device in more challenging conditions, such as running, climbing stairs, or other tasks requiring greater mechanical effort, remains unexplored. Future studies should therefore assess the effects of the exoskeleton in these higher-demanding scenarios to better characterise and understand its assistance capabilities. 5. Conclusions In this study, we investigated the interaction between the Hypershell-X exoskeleton and a cohort of healthy subjects during overground walking through a comprehensive kinematic analysis. Overall, the device demonstrated a good level of preservation of the physiological gait, as both spatiotemporal parameters and joint kinematics were maintained across conditions. Only at the highest assistance level tested did we observe significant alterations in hip RoM and skewness, but the angular patterns remained strongly correlated with physiological gait. These findings suggest that the assistance delivered by the exoskeleton is well integrated into the user’s natural movement, although excessive support may lead to localized modifications at the hip. Future studies should focus on fully characterising the underlying motor control strategies during the interaction with the device by combining kinematic data with EMG-based analyses, such as muscle synergies. Moreover, evaluating the device in more demanding tasks will help explore its full assistance potential. Declarations Author contributions NM carried out experiments, analysed data, wrote and reviewed the paper. VL carried out experiments, analysed data, wrote and reviewed the paper. CB carried out experiments, analysed data, wrote and reviewed the paper. LMT reviewed the paper and was responsible for the supervision and acquisition of the funding. AS carried out experiments, analysed the data, wrote and reviewed the paper, participated in the conceptualization of the study and supervised the work. All authors have read and agreed to the published version of the manuscript. Fundings This work is supported by the Italian Ministry of University and Research, under the complementary actions to the Plan of National Recovery and Resilience (PNRR) “Fit4MedRob -Fit for Medical Robotics” Grant (PNC0000007). Data availability The dataset obtained during the study is available from the corresponding author upon reasonable request. Ethical statement This study was reviewed and approved by the Ethics Committee of the National Research Council (approval number N°231878/2025, approved on 24/06/2025), according to the Declaration of Helsinki. References Bogue, R. (2015). Robotic Exoskeletons: A Review of Recent Progress. Industrial Robot , 42 (1), 5–10. https://doi.org/10.1108/IR-08-2014-0379 Bryan, G. M., Franks, P. W., Song, S., Voloshina, A. S., Reyes, R., O’Donovan, M. P., Gregorczyk, K. N., & Collins, S. H. (2021). Optimized hip-knee-ankle exoskeleton assistance at a range of walking speeds. Journal of Neuroengineering and Rehabilitation , 18 (1), 152. https://doi.org/10.1186/s12984-021-00943-y Calabrò, R. S., Sorrentino, G., Cassio, A., Mazzoli, D., Andrenelli, E., Bizzarini, E., Campanini, I., Carmignano, S. M., Cerulli, S., Chisari, C., Colombo, V., Dalise, S., Fundarò, C., Gazzotti, V., Mazzoleni, D., Mazzucchelli, M., Melegari, C., Merlo, A., Stampacchia, G., … Italian Consensus Conference on Robotics in Neurorehabilitation (CICERONE). (2021). Robotic-assisted gait rehabilitation following stroke: A systematic review of current guidelines and practical clinical recommendations. European Journal of Physical and Rehabilitation Medicine , 57 (3), 460–471. https://doi.org/10.23736/S1973-9087.21.06887-8 Calafiore, D., Negrini, F., Tottoli, N., Ferraro, F., Ozyemisci-Taskiran, O., & de Sire, A. (2022). Efficacy of robotic exoskeleton for gait rehabilitation in patients with subacute stroke: A systematic review. European Journal of Physical and Rehabilitation Medicine , 58 (1), 1–8. https://doi.org/10.23736/S1973-9087.21.06846-5 Camardella, C., Porcini, F., Filippeschi, A., Marcheschi, S., Solazzi, M., & Frisoli, A. (2021). Gait Phases Blended Control for Enhancing Transparency on Lower-Limb Exoskeletons. IEEE Robotics and Automation Letters , 6 (3), 5453–5460. https://doi.org/10.1109/LRA.2021.3075368 Cardoso, A., Ribeiro, A., Carneiro, P., & Colim, A. (2024). Evaluating Exoskeletons for WMSD Prevention: A Systematic Review of Applications and Ergonomic Approach in Occupational Settings. International Journal of Environmental Research and Public Health , 21 (12), Article 12. https://doi.org/10.3390/ijerph21121695 Carpinella, I., Cattaneo, D., Abuarqub, S., & Ferrarin, M. (2009). Robot-based rehabilitation of the upper limbs in multiple sclerosis: Feasibility and preliminary results. Journal of Rehabilitation Medicine , 41 (12), 966–970. https://doi.org/10.2340/16501977-0401 Chen, W., Wu, S., Zhou, T., & Xiong, C. (2018). On the biological mechanics and energetics of the hip joint muscle-tendon system assisted by passive hip exoskeleton. Bioinspiration & Biomimetics , 14 (1), 016012. https://doi.org/10.1088/1748-3190/aaeefd Cherni, Y., Blache, Y., Begon, M., Ballaz, L., & Dal Maso, F. (2023). Effect of Robotic-Assisted Gait at Different Levels of Guidance and Body Weight Support on Lower Limb Joint Kinematics and Coordination. Sensors (Basel, Switzerland) , 23 (21), 8800. https://doi.org/10.3390/s23218800 d’Avella, A., Portone, A., Fernandez, L., & Lacquaniti, F. (2006). Control of fast-reaching movements by muscle synergy combinations. Journal of Neuroscience , 26 (30), 7791–7810. https://doi.org/10.1523/JNEUROSCI.0830-06.2006 de Looze, M. P., Bosch, T., Krause, F., Stadler, K. S., & O’Sullivan, L. W. (2016). Exoskeletons for Industrial Application and Their Potential Effects on Physical Work Load. Ergonomics , 59 (5), 671–681. https://doi.org/10.1080/00140139.2015.1081988 Di Tommaso, F., Tamburella, F., Lorusso, M., Gastaldi, L., Molinari, M., & Tagliamonte, N. L. (2023). Biomechanics of Exoskeleton-Assisted Treadmill Walking. 2023 International Conference on Rehabilitation Robotics (ICORR) , 1–6. https://doi.org/10.1109/ICORR58425.2023.10304685 Ding, Y., Kim, M., Kuindersma, S., & Walsh, C. J. (2018). Human-in-the-loop optimization of hip assistance with a soft exosuit during walking. Science Robotics , 3 (15), eaar5438. https://doi.org/10.1126/scirobotics.aar5438 Fraiwan, L., & Hassanin, O. (2021). Computer-aided identification of degenerative neuromuscular diseases based on gait dynamics and ensemble decision tree classifiers. PLoS ONE , 16 (6), e0252380. https://doi.org/10.1371/journal.pone.0252380 Fritz, H., Patzer ,Diane, & and Galen, S. S. (2019). Robotic exoskeletons for reengaging in everyday activities: Promises, pitfalls, and opportunities. Disability and Rehabilitation , 41 (5), 560–563. https://doi.org/10.1080/09638288.2017.1398786 Gassert, R., & Dietz, V. (2018). Rehabilitation robots for the treatment of sensorimotor deficits: A neurophysiological perspective. Journal of NeuroEngineering and Rehabilitation , 15 , 46. https://doi.org/10.1186/s12984-018-0383-x Grimmer, M., & Zhao, G. (2024). Hip Exoskeleton for Cycling Assistance. Bioengineering , 11 (7), 683. https://doi.org/10.3390/bioengineering11070683 Hsu, T.-H., Tsai, C.-L., Chi, J.-Y., Hsu, C.-Y., & Lin, Y.-N. (2023). Effect of wearable exoskeleton on post-stroke gait: A systematic review and meta-analysis. Annals of Physical and Rehabilitation Medicine , 66 (1), 101674. https://doi.org/10.1016/j.rehab.2022.101674 Ishmael, M. K., Gunnell, A., Pruyn, K., Creveling, S., Hunt, G., Hood, S., Archangeli, D., Foreman, K. B., & Lenzi, T. (2023). Powered Hip Exoskeleton Reduces Residual Hip Effort Without Affecting Kinematics and Balance in Individuals With Above-Knee Amputations During Walking. IEEE Transactions on Bio-Medical Engineering , 70 (4), 1162–1171. https://doi.org/10.1109/TBME.2022.3211842 Ivanenko, Y. P., Poppele, R. E., & Lacquaniti, F. (2004). Five basic muscle activation patterns account for muscle activity during human locomotion. The Journal of Physiology , 556 (Pt 1), 267–282. https://doi.org/10.1113/jphysiol.2003.057174 Lee, Y., Roh, S., Lee, M., Choi, B., Lee, J., Kim, J., Choi, H., Shim, Y., & Kim, Y.-J. (2017). A flexible exoskeleton for hip assistance. 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , 1058–1063. https://doi.org/10.1109/IROS.2017.8202275 Lim, B., Lee, J., Jang, J., Kim, K., Park, Y. J., Seo, K., & Shim, Y. (2019). Delayed Output Feedback Control for Gait Assistance With a Robotic Hip Exoskeleton. Trans. Rob. , 35 (4), 1055–1062. https://doi.org/10.1109/TRO.2019.2913318 Marimon, X., Mengual, I., López-de-Celis, C., Portela, A., Rodríguez-Sanz, J., Herráez, I. A., & Pérez-Bellmunt, A. (2024). Kinematic Analysis of Human Gait in Healthy Young Adults Using IMU Sensors: Exploring Relevant Machine Learning Features for Clinical Applications. Bioengineering (Basel, Switzerland) , 11 (2), 105. https://doi.org/10.3390/bioengineering11020105 Moscatelli, N., Brambilla, C., Lanzani, V., Molinari Tosatti, L., & Scano, A. (2025). Assessing Transparency of Robots, Exoskeletons, and Assistive Devices: A Systematic Review. Sensors , 25 (14), 4444. https://doi.org/10.3390/s25144444 Nepomuceno, P., Souza, W. H., Pakosh, M., Musselman, K. E., & Craven, B. C. (2024). Exoskeleton-based exercises for overground gait and balance rehabilitation in spinal cord injury: A systematic review of dose and dosage parameters. Journal of Neuroengineering and Rehabilitation , 21 (1), 73. https://doi.org/10.1186/s12984-024-01365-2 Nurse, C. A., Wolf, D. N., Rodzak, K. M., Teater, R. H., Ice, C. C., Fine, S. J., Holtzman, E. C., & Zelik, K. E. (2025). Evaluating the Biomechanical Effects and Real-World Usability of a Novel Ankle Exo for Runners. Journal of Biomechanical Engineering , 147 (3), 031004. https://doi.org/10.1115/1.4067579 Park, Y.-H., Lee, D.-H., & Lee, J.-H. (2024). A Comprehensive Review: Robot-Assisted Treatments for Gait Rehabilitation in Stroke Patients. Medicina (Kaunas, Lithuania) , 60 (4), 620. https://doi.org/10.3390/medicina60040620 Proietti, T., Crocher, V., Roby-Brami, A., & Jarrasse, N. (2016). Upper-Limb Robotic Exoskeletons for Neurorehabilitation: A Review on Control Strategies. IEEE Reviews in Biomedical Engineering , 9 , 1–1. https://doi.org/10.1109/RBME.2016.2552201 Sanz-Morère, C. B., Martini, E., Meoni, B., Arnetoli, G., Giffone, A., Doronzio, S., Fanciullacci, C., Parri, A., Conti, R., Giovacchini, F., Friðriksson, Þ., Romo, D., Crea, S., Molino-Lova, R., & Vitiello, N. (2021). Robot-mediated overground gait training for transfemoral amputees with a powered bilateral hip orthosis: A pilot study. Journal of NeuroEngineering and Rehabilitation , 18 (1), 111. https://doi.org/10.1186/s12984-021-00902-7 Sawicki, G. S., Beck, O. N., Kang, I., & Young, A. J. (2020). The exoskeleton expansion: Improving walking and running economy. Journal of Neuroengineering and Rehabilitation , 17 (1), 25. https://doi.org/10.1186/s12984-020-00663-9 Seo, K., Lee, J., Lee, Y., Ha, T., & Shim, Y. (2016). Fully autonomous hip exoskeleton saves metabolic cost of walking. 2016 IEEE International Conference on Robotics and Automation (ICRA) , 4628–4635. https://doi.org/10.1109/ICRA.2016.7487663 Srivatsa, A. V., Aghamohammadi, N. R., Ryali, P., Scarpellini, A., Kim, J. J. H., Aggarwal, A., Wright, Z., Huang, F. C., Reinkensmeyer, D. J., Esmailbeigi, H., & Patton, J. L. (2025). Distribution Analysis for Diagnostics and Therapeutics of Motor Actions. IEEE Journal of Biomedical and Health Informatics , PP . https://doi.org/10.1109/JBHI.2025.3636009 Stramel, D., & Agrawal, S. (2022). Assessing Changes in Human Gait with a Mobile Tethered Pelvic Assist Device (mTPAD) in Transparent Mode with Hand Holding Conditions. Proceedings of the IEEE International Conference on Biomedical Robotics and Biomechatronics (BioRob) , 1–6. https://doi.org/10.1109/BioRob52689.2022.9925359 Yandell, M. B., Tacca, J. R., & Zelik, K. E. (2019). Design of a Low Profile, Unpowered Ankle Exoskeleton That Fits Under Clothes: Overcoming Practical Barriers to Widespread Societal Adoption. IEEE Transactions on Neural Systems and Rehabilitation Engineering , 27 (4), 712–723. https://doi.org/10.1109/TNSRE.2019.2904924 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-8367199","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":560627787,"identity":"6114e5de-5fdf-4e24-b17c-d630fed54f47","order_by":0,"name":"Nicol Moscatelli","email":"","orcid":"","institution":"Institute of Intelligent Industrial Technologies and Systems for Advanced Manufacturing (STIIMA), Italian National Research Council (CNR), Milano, Italy)","correspondingAuthor":false,"prefix":"","firstName":"Nicol","middleName":"","lastName":"Moscatelli","suffix":""},{"id":560627788,"identity":"b523ab2a-bf65-4923-9f0a-96c5c2f23747","order_by":1,"name":"Valentina Lanzani","email":"","orcid":"","institution":"Institute of Intelligent Industrial Technologies and Systems for Advanced Manufacturing (STIIMA), Italian National Research Council (CNR), Milano, Italy)","correspondingAuthor":false,"prefix":"","firstName":"Valentina","middleName":"","lastName":"Lanzani","suffix":""},{"id":560628089,"identity":"f7296380-ba5a-4603-a38c-345b4240f45e","order_by":2,"name":"Cristina Brambilla","email":"","orcid":"","institution":"Institute of Intelligent Industrial Technologies and Systems for Advanced Manufacturing (STIIMA), Italian National Research Council (CNR), Milano, Italy)","correspondingAuthor":false,"prefix":"","firstName":"Cristina","middleName":"","lastName":"Brambilla","suffix":""},{"id":560628090,"identity":"97d780c0-ac20-4c45-bff1-0b215ebc53e2","order_by":3,"name":"Lorenzo Molinari Tosatti","email":"","orcid":"","institution":"Institute of Intelligent Industrial Technologies and Systems for Advanced Manufacturing (STIIMA), Italian National Research Council (CNR), Milano, Italy)","correspondingAuthor":false,"prefix":"","firstName":"Lorenzo","middleName":"Molinari","lastName":"Tosatti","suffix":""},{"id":560628091,"identity":"d14b682a-2849-4f78-a5cb-54d9b4bf32dc","order_by":4,"name":"Alessandro Scano","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA50lEQVRIiWNgGAWjYDACdgjF2AAT4AeTBni0MKNrkQSzDPDowdBicICANfzNzIc//qi5I7u9vYH5A5AhZ3z87LEPDAV/cGqROMyWJs1z7JnxnDMH2CR7gAyzM3nJM/A67DCPGTMD2+HEGRIJbGDGths8xnj9In+Y//PHH//AWpg/MwAZm2cQ0GJwmIdBgrcNrIVBmhHI2CBBQIvhYTYzad6+w8YzeA62SfYCGRJncowZEgyMcWqRO978+OOPb4dlZ7A3H/4AZMjxt58xZvjwRw639xEAETUMDAnEaBgFo2AUjIJRgBMAACX/UITxwMPoAAAAAElFTkSuQmCC","orcid":"","institution":"Institute of Intelligent Industrial Technologies and Systems for Advanced Manufacturing (STIIMA), Italian National Research Council (CNR), Milano, Italy)","correspondingAuthor":true,"prefix":"","firstName":"Alessandro","middleName":"","lastName":"Scano","suffix":""}],"badges":[],"createdAt":"2025-12-15 14:17:19","currentVersionCode":1,"declarations":{"humanSubjects":true,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":true,"humanSubjectConsent":true,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-8367199/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8367199/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":98318153,"identity":"3ca3766e-87a1-41d2-afa8-cc6adf3dd4ff","added_by":"auto","created_at":"2025-12-16 13:40:06","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10956460,"visible":true,"origin":"","legend":"","description":"","filename":"ImpactofHipExoskeletonAssistanceonHumanGait.docx","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/f1b20f9f9a14ccfd285d9ecc.docx"},{"id":98318148,"identity":"d442e39d-91e8-488b-8170-8cb6f605ee6f","added_by":"auto","created_at":"2025-12-16 13:40:05","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":342,"visible":true,"origin":"","legend":"","description":"","filename":"rs8367199.json","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/61ae04472023b8c6c88559ad.json"},{"id":98318154,"identity":"048980ce-278b-4351-8922-7f60f0136a27","added_by":"auto","created_at":"2025-12-16 13:40:06","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":104005,"visible":true,"origin":"","legend":"","description":"","filename":"rs83671990enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/157b0b4f08ffe006395f5fba.xml"},{"id":98436827,"identity":"e8f81df9-f59d-4db7-b7a3-1fd4b3181dcb","added_by":"auto","created_at":"2025-12-17 16:56:18","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6000558,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/7aef85931604a2ceea93f099.png"},{"id":98437248,"identity":"d0604d04-819a-4760-ae4a-b9f8d372b9f1","added_by":"auto","created_at":"2025-12-17 16:57:10","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":717972,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/9ddeef61eb8310518b53e4e8.png"},{"id":98318167,"identity":"f3e4798f-76fc-4588-b980-388b4c7b2360","added_by":"auto","created_at":"2025-12-16 13:40:07","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":170053,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/fe3cc9eed9d8fa97c3322a46.png"},{"id":98436077,"identity":"c545a4cf-4465-4120-bf9e-2f3db4acf73f","added_by":"auto","created_at":"2025-12-17 16:54:52","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":278010,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/2ec9da8ce9cbfccd9b17a4c2.png"},{"id":98318156,"identity":"7da0cf17-e0b8-410b-b148-8eb515496700","added_by":"auto","created_at":"2025-12-16 13:40:07","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":315306,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/fa2cfb4df45f14983b5a6450.png"},{"id":98318161,"identity":"07d47061-91bf-49b1-89e1-487fab01b726","added_by":"auto","created_at":"2025-12-16 13:40:07","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":666533,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/5312d1e66d3b5281c4926f42.png"},{"id":98436214,"identity":"bc10e0f5-6937-4ffe-8311-2c2e4811398d","added_by":"auto","created_at":"2025-12-17 16:55:06","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":152653,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/e66dd7070ed1f00618a82e34.png"},{"id":98436833,"identity":"a4e92787-17e5-499a-9bbc-257de09bd535","added_by":"auto","created_at":"2025-12-17 16:56:19","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":61331,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/3b5867da34d4a5953dc6c3bd.png"},{"id":98318162,"identity":"cc17d885-8f7d-40c1-b394-3e27ba7ebe0a","added_by":"auto","created_at":"2025-12-16 13:40:07","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":49491,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/200c2683f06f4814d4cecd3e.png"},{"id":98318151,"identity":"3e11c827-2af3-4f57-9cc3-202f95cba322","added_by":"auto","created_at":"2025-12-16 13:40:06","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":56703,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/9bbc2d9e20b5b31f8f108cac.png"},{"id":98318166,"identity":"a9dc097d-0e3b-4256-a6f9-de90dd48c4b1","added_by":"auto","created_at":"2025-12-16 13:40:07","extension":"xml","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":101840,"visible":true,"origin":"","legend":"","description":"","filename":"rs83671990structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/84a640063c94bf250932eff4.xml"},{"id":98437284,"identity":"080baedd-e40d-40ca-a89c-2f4af0858faa","added_by":"auto","created_at":"2025-12-17 16:57:10","extension":"html","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":111366,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/b7397f86670bfbb8c6496006.html"},{"id":98318155,"identity":"2352cbbc-24ab-480f-a648-05a47eb828f3","added_by":"auto","created_at":"2025-12-16 13:40:07","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":130900,"visible":true,"origin":"","legend":"\u003cp\u003eInstrumentation setup showing the Hypershell X exoskeleton and IMU placement used to capture lower-limb kinematics during walking trials.\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/dfd3346fa116cd0dace995b5.jpg"},{"id":98318157,"identity":"d1906386-186b-41d8-9be6-96d9cda478a8","added_by":"auto","created_at":"2025-12-16 13:40:07","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":111770,"visible":true,"origin":"","legend":"\u003cp\u003eJoint angle trajectories for a representative participant, shown as mean ± standard deviation of the gait cycle phase. The Free condition was used as the reference to compare the other three modalities. In each plot, solid, coloured lines indicate the mean joint angle, while the shaded areas represent the corresponding standard deviations. Coloured, dashed lines mark the toe-off event, separating the stance from the swing phase, and each line matches the colour of its corresponding condition. Hip FE = hip flexion-extension, Knee FE = knee flexion-extension, Ankle DP = ankle dorsiflexion-plantarflexion. F = Free, in green; T = Transparent, in orange; A1 = Assisted Level 1, in purple; A3 = Assisted Level 3, in pink.\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/22f3265134e8d5ea7f9233dc.jpg"},{"id":98437421,"identity":"ec1ff965-cb38-4f17-89d5-5946ae0e6d1f","added_by":"auto","created_at":"2025-12-17 16:57:18","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":69508,"visible":true,"origin":"","legend":"\u003cp\u003eHeatmap of correlation coefficients (r) between walking conditions for each joint. All values are above 0.9, indicating a good level of similarity (p \u0026lt; 0.001). Hip FE = hip flexion-extension, Knee FE = knee flexion-extension, Ankle DP = ankle dorsiflexion-plantarflexion.\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/ec381e30cf4911c8f4e43b43.jpg"},{"id":98318164,"identity":"b81e82ad-fb53-453f-a760-b3ab9ea38009","added_by":"auto","created_at":"2025-12-16 13:40:07","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":78550,"visible":true,"origin":"","legend":"\u003cp\u003eHip (top left), knee (top right), and ankle (bottom) flexion-extension RoM across the four experimental conditions (F = Free, in green; T = Transparent, in orange; A1 = Assisted Level 1, in purple; A3 = Assisted Level 3, in pink). For each joint, the raincloud plots showing the individual behaviour of each subject are reported, along with boxplots summarizing within-condition variability. The distributions of data are also presented to illustrate their normality. In the raincloud plots, each subject is represented by a single dot, and grey lines connect measurements from the same subject across conditions. Hip FE = hip flexion-extension, Knee FE = knee flexion-extension, Ankle DP = ankle dorsiflexion-plantarflexion. * p \u0026lt; 0.05, ** p \u0026lt; 0.01.\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/001df0f75e3c3623a70e48d1.jpg"},{"id":98318150,"identity":"383fc197-8351-4781-8191-f5bb817bad7c","added_by":"auto","created_at":"2025-12-16 13:40:05","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":89432,"visible":true,"origin":"","legend":"\u003cp\u003eSkewness of hip (top left), knee (top right), and ankle (bottom) flexion-extension angle across the four experimental conditions. (F = Free, in green; T = Transparent, in orange; A1 = Assisted Level 1, in purple; A3 = Assisted Level 3, in pink). For each joint, the raincloud plots showing the individual behaviour of each subject are reported, along with boxplots summarizing within-condition variability. The distributions of data are also presented to illustrate their normality. In the raincloud plots, each subject is represented by a single dot, and grey lines connect measurements from the same subject across conditions. Hip FE = hip flexion-extension, Knee FE = knee flexion-extension, Ankle DP = ankle dorsiflexion-plantarflexion. * p \u0026gt; 0.05, ** p \u0026lt; 0.01.\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/f936537c2a4e06efc58ab7a9.jpg"},{"id":98774789,"identity":"d94885c7-a0e8-4109-96c8-a93130260a98","added_by":"auto","created_at":"2025-12-22 12:14:20","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1117956,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8367199/v1/04c23018-a454-433a-94dc-1708a13aa9eb.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eImpact of Hip Exoskeleton Assistance on Human Gait: A Kinematic Study\u003c/p\u003e","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eWearable lower-limb exoskeletons have emerged as a versatile technology for a wide range of applications.\u003c/p\u003e \u003cp\u003eIn clinical settings, they are widely used to promote and support motor recovery in individuals with neurological impairments (Carpinella et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Fritz et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Gassert \u0026amp; Dietz, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Nepomuceno et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Park et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), and many studies have demonstrated that exoskeleton-based rehabilitation can be comparable, or even superior, to conventional physiotherapy treatments (Calabr\u0026ograve; et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Calafiore et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Hsu et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Moreover, lower-limb exoskeletons have been used for gait training and daily assistance in individuals with above-knee amputations (Ishmael et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Sanz-Mor\u0026egrave;re et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), who often exhibit abnormal gait patterns and compensatory movement strategies. Beyond rehabilitation, exoskeletons are increasingly adopted in non-clinical contexts. Industrial wearable devices, for example, can support workers in physically demanding tasks, helping to reduce fatigue and the risk of developing musculoskeletal disorders (Bogue, \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Cardoso et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; de Looze et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). More recently, interest has also grown in sports and recreational applications, where wearable assistive technologies aim to reduce energy expenditure (Grimmer \u0026amp; Zhao, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Sawicki et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), improve performance (Yandell et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), and prevent injuries (Nurse et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eLower-limb exoskeletons typically provide assistance at the hip, at the ankle, or at multiple joints (Bryan et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Among these configurations, hip exoskeletons have shown consistent reductions in the metabolic cost of walking in healthy individuals compared to unassisted gait (Ding et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Lee et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Lim et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Seo et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), and these reductions are often larger than those reported with ankle exoskeletons (Sawicki et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). A key factor contributing to this aspect is the location of the device\u0026rsquo;s added mass: hip exoskeletons position most of their components close to the trunk and near the body\u0026rsquo;s center of mass, which helps minimize negative effects and disturbances to natural gait biomechanics (Chen et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Ishmael et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Sawicki et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). For these reasons, hip exoskeletons are emerging as promising solutions for improving walking efficiency while preserving comfort and natural movement.\u003c/p\u003e \u003cp\u003eWithin this context, the Hypershell-X exoskeleton (Hypershell, Shanghai, China) is a lightweight (2.0 Kg) hip exoskeleton designed for assisting a large variety of recreational tasks, including gait, cycling, stairs ascent and descent, mountain climbing, running, and fast walking, with the goal of reducing physical effort by 30%.\u003c/p\u003e \u003cp\u003eAs these technologies continue to evolve, understanding how exoskeleton assistance influences human movement has become essential for optimizing human-robot interaction. In this regard, kinematic analysis plays a crucial role: while metrics such as metabolic cost are valuable for assessing the effectiveness of the assistance, a complete kinematic analysis provides deeper insights into the quality of the interaction with the device. Spatiotemporal parameters such as cadence and walking speed, combined with joint angle trajectories and coordination patterns, reflect whether the exoskeleton supports the user without compromising the physiological characteristics of gait.\u003c/p\u003e \u003cp\u003eIn this context, the present pilot study evaluates the interaction between the Hypershell-X exoskeleton and healthy adults during overground walking. By performing a detailed kinematic analysis, we aim to characterize how different levels of robotic assistance modulate human movement, coordination, and functional performance.\u003c/p\u003e"},{"header":"2. Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\n \u003ch2\u003e2.1. Equipment\u003c/h2\u003e\n \u003cp\u003eThe Hypershell-X exoskeleton is an AI-powered wearable device designed to augment human mobility and reduce fatigue during recreational activities. Its control system detects movement patterns in real time and provides adaptive power assistance. The device integrates a motor with 800W peak power and 32Nm torque, delivering substantial support tailored to the specific movement being performed. The Hypershell X includes several assistance modes, each optimized for different activities. The \u0026ldquo;Eco\u0026rdquo; mode is the default setting and is programmed to support moderate activities such as walking or hiking. The \u0026ldquo;Hyper\u0026rdquo; mode, instead, offers more dynamic assistance for higher intensity tasks, such as running or climbing. Moreover, a \u0026ldquo;Transparent\u0026rdquo; mode is also available, allowing the user to temporarily suspend assistance. Both the \u0026ldquo;Eco\u0026rdquo; and \u0026ldquo;Hyper\u0026rdquo; modes offer four adjustable assistance levels, enabling the user to tailor the amount of support to the task intensity.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\n \u003ch2\u003e2.2. Participants\u003c/h2\u003e\n \u003cp\u003eA total of 16 healthy participants (9 males, mean age 37.6\u0026thinsp;\u0026plusmn;\u0026thinsp;13 years, mean height 180\u0026thinsp;\u0026plusmn;\u0026thinsp;6.5 cm, mean weight 74.4\u0026thinsp;\u0026plusmn;\u0026thinsp;6.8 kg; 7 females, mean age 26.1\u0026thinsp;\u0026plusmn;\u0026thinsp;1.7 years, mean height 165\u0026thinsp;\u0026plusmn;\u0026thinsp;5.3 cm, mean weight 51.3\u0026thinsp;\u0026plusmn;\u0026thinsp;3.4 kg) were recruited for this study. None of the participants reported any lower-limb injuries, surgical interventions, neurological or musculoskeletal disorders that could affect their gait. All subjects except two were right-limb dominant, which was determined by their preferred foot used to initiate walking. Before participation, all individuals provided written informed consent in accordance with protocols approved by the Ethics Committee of the National Research Council (approval number N\u0026deg;231878/2025, approved on 24/06/2025) and in compliance with the Declaration of Helsinki.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\n \u003ch2\u003e2.3. Experimental procedure and data recording\u003c/h2\u003e\n \u003cp\u003eAll participants performed level-ground walking along a 10-meter straight path at a self-selected speed while maintaining a level head posture, as during normal walking. Subjects started from a designated point on the floor and walked to a marker 10 meters away, then stopped, turned around, and repeated the task, completing a total of 6 trials. Walking sessions were repeated under four experimental conditions:\u003c/p\u003e\u003cspan\u003e\n \u003cp\u003e\u003cem\u003ea. Free\u003c/em\u003e (F): walking without the exoskeleton\u003c/p\u003e\n \u003c/span\u003e \u003cspan\u003e\n \u003cp\u003e\u003cem\u003eb. Transparent\u003c/em\u003e (T): walking while wearing Hypershell-X, with no active assistance\u003c/p\u003e\n \u003c/span\u003e \u003cspan\u003e\n \u003cp\u003e\u003cem\u003ec. Assisted\u003c/em\u003e: walking with active assistance provided by the exoskeleton. Two different assistance levels were tested, low (level 1 \u0026ndash; A1) and high (level 3 \u0026ndash; A3), for the \u0026ldquo;Eco\u0026rdquo; mode of the Hypershell-X. The A3 level was intentionally included, even though it was suggested for more demanding activities, with the aim of evaluating how physiological gait is affected when providing a high level of assistance.\u003c/p\u003e\n \u003c/span\u003e\n \u003cp\u003eData acquisition was carried out in the corridor located in front of the Bio-SimPro-Lab (Advanced Methods for Biomedical Signal and Image Processing Laboratory) at the National Research Council in Milan, Italy. Kinematic data were collected using seven inertial measurement units (IMUs) provided by Captiks S.r.l. (Rome, Italy). One IMU was placed around the waist, and three sensors were positioned on each leg at the foot, ankle and thigh (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). This configuration enabled the reconstruction of the three degrees of freedom of the pelvis and of the hip, knee and ankle joints. Electromyographic (EMG) signals were also recorded during the trials; however, EMG data were not included in the analysis presented in this study.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\n \u003ch2\u003e2.4. Data processing and outcome metrics\u003c/h2\u003e\n \u003cp\u003eJoint angles obtained from the IMUs were pre-processed using a 5th -order, low-pass Butterworth filter, with a cutting frequency of 10 Hz. The segmentation of the recordings into individual gait cycles was performed using the Captiks Motion Analyzer software, which also allowed the extraction of the spatiotemporal gait parameters. Specifically, the following parameters were considered: cadence (steps/min); walking speed (m/s); normalized walking speed (%/s), obtained by normalizing walking speed by the participant\u0026rsquo;s height; step length (m), defined as the distance between the heel strike of one foot and the subsequent heel strike of the opposite foot; stride length (m), defined as the distance between two consecutive heel strikes of the same foot; stance time (s), defined as the time interval between heel strike and toe-off; swing time (s), defined as the time interval between the toe-off and the next heel strike; stride time (s), defined as the duration of the full gait cycle.\u003c/p\u003e\n \u003cp\u003eSince only healthy individuals were recruited, gait symmetry between limbs was assumed, and therefore we performed the analysis using data from the dominant limb only.\u003c/p\u003e\n \u003cp\u003eIn addition to the spatiotemporal parameters, hip, knee, and ankle flexion-extension angles were used to compute the Range of Motion (RoM) for each joint, defined as the difference between the maximum and minimum angles within the gait cycle. To further assess potential changes in gait patterns across different assistance modes, skewness was computed from the joint angle waveforms. Skewness quantifies the temporal asymmetry of a signal around its mean (Srivatsa et al., \u003cspan class=\"CitationRef\"\u003e2025\u003c/span\u003e). Negative values indicate that the signal remains predominantly below its mean, which, for joint angles, may reflect a greater proportion of time spent in a joint extension phase. Conversely, positive values indicate that the signal remains predominantly above the mean, suggesting more time spent in flexion. Values close to zero reflect a more balanced, symmetric distribution around the mean (Fraiwan \u0026amp; Hassanin, \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e; Marimon et al., \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e). Skewness was computed using the \u003cem\u003eskewness\u003c/em\u003e built-in function in MATLAB (MathWorks, Natick, USA).\u003c/p\u003e\n \u003cp\u003eAll variables were first tested for normality using the Shapiro-Wilk test. Since all variables met the normality assumption, a repeated-measures ANOVA was performed, with the assistance condition (Free Movement (F), Transparent (T), Assistance Level 1 (A1), Assistance Level 3 (A3)) considered as the main factor. In addition, Spearman\u0026rsquo;s correlation coefficient was computed between joint angle waveforms across conditions to assess the similarity and coherence of gait patterns. Statistical significance was set at p\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"3. Results","content":"\u003cp\u003eFigure 2 illustrates the joint angles on the sagittal plane for a representative subject. Although the mean pattern was consistent across all participants, we chose to present the data from a single subject to provide a cleaner and more readable visualization. This choice also makes it easier to highlight relevant gait events, such as the toe-off instants across conditions.\u003c/p\u003e\n\u003cp\u003eJoint angle trajectories for a representative participant, shown as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation of the gait cycle phase. The \u003cem\u003eFree\u003c/em\u003e condition was used as the reference to compare the other three modalities. In each plot, solid, coloured lines indicate the mean joint angle, while the shaded areas represent the corresponding standard deviations. Coloured, dashed lines mark the toe-off event, separating the stance from the swing phase, and each line matches the colour of its corresponding condition. Hip FE\u0026thinsp;=\u0026thinsp;hip flexion-extension, Knee FE\u0026thinsp;=\u0026thinsp;knee flexion-extension, Ankle DP\u0026thinsp;=\u0026thinsp;ankle dorsiflexion-plantarflexion. F\u0026thinsp;=\u0026thinsp;\u003cem\u003eFree\u003c/em\u003e, in green; T\u0026thinsp;=\u0026thinsp;\u003cem\u003eTransparent\u003c/em\u003e, in orange; A1\u0026thinsp;=\u0026thinsp;\u003cem\u003eAssisted Level 1\u003c/em\u003e, in purple; A3\u0026thinsp;=\u0026thinsp;\u003cem\u003eAssisted Level 3\u003c/em\u003e, in pink.\u003c/p\u003e\n\u003cp\u003eThe correlation between the sagittal angle profiles of the main joints (hip, knee, and ankle) across the four different walking conditions (F, T, A1, and A3) was very high (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e). The correlation values between conditions, averaged across all subjects, were all above 0.9 (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), indicating a strong similarity and preservation of joint movement patterns despite the different levels of assistance provided by the exoskeleton.\u003c/p\u003e\n\u003cp\u003eThe mean values for the spatiotemporal parameters across the four conditions are reported in Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e. No significant differences were found, indicating that the overall gait pattern was not altered by the different levels of assistance provided.\u003c/p\u003e\n\u003cdiv class=\"gridtable\"\u003e\u0026nbsp;\u003ctable id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eSpatiotemporal parameters across the four walking conditions (averaged between subjects). Each value is expressed as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation across all participants.\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"5\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eFree\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eTransparent\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eAssisted 1\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eAssisted 3\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eCadence (steps/min)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e99.2\u0026thinsp;\u0026plusmn;\u0026thinsp;8.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e99.6\u0026thinsp;\u0026plusmn;\u0026thinsp;9.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e99.8\u0026thinsp;\u0026plusmn;\u0026thinsp;9.86\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e101.6\u0026thinsp;\u0026plusmn;\u0026thinsp;12.11\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eWalking speed (m/s)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.01\u0026thinsp;\u0026plusmn;\u0026thinsp;0.14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.96\u0026thinsp;\u0026plusmn;\u0026thinsp;0.15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.02\u0026thinsp;\u0026plusmn;\u0026thinsp;0.16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.03\u0026thinsp;\u0026plusmn;\u0026thinsp;0.21\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eNormalized walking speed (%/s)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e58.4\u0026thinsp;\u0026plusmn;\u0026thinsp;7.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e55.4\u0026thinsp;\u0026plusmn;\u0026thinsp;7.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e58.9\u0026thinsp;\u0026plusmn;\u0026thinsp;9.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e59.5\u0026thinsp;\u0026plusmn;\u0026thinsp;12.65\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eStep length (m)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.60\u0026thinsp;\u0026plusmn;\u0026thinsp;0.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.56\u0026thinsp;\u0026plusmn;\u0026thinsp;0.09\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.59\u0026thinsp;\u0026plusmn;\u0026thinsp;0.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.60\u0026thinsp;\u0026plusmn;\u0026thinsp;0.09\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eStride length (m)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.23\u0026thinsp;\u0026plusmn;\u0026thinsp;0.14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.16\u0026thinsp;\u0026plusmn;\u0026thinsp;0.15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.22\u0026thinsp;\u0026plusmn;\u0026thinsp;0.14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.21\u0026thinsp;\u0026plusmn;\u0026thinsp;0.16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eStance time (s)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.75\u0026thinsp;\u0026plusmn;\u0026thinsp;0.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.74\u0026thinsp;\u0026plusmn;\u0026thinsp;0.09\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.74\u0026thinsp;\u0026plusmn;\u0026thinsp;0.10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.72\u0026thinsp;\u0026plusmn;\u0026thinsp;0.10\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eSwing time (s)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.48\u0026thinsp;\u0026plusmn;\u0026thinsp;0.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.48\u0026thinsp;\u0026plusmn;\u0026thinsp;0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.47\u0026thinsp;\u0026plusmn;\u0026thinsp;0.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.48\u0026thinsp;\u0026plusmn;\u0026thinsp;0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eStride time (s)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.22\u0026thinsp;\u0026plusmn;\u0026thinsp;0.11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.22\u0026thinsp;\u0026plusmn;\u0026thinsp;0.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.22\u0026thinsp;\u0026plusmn;\u0026thinsp;0.13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.20\u0026thinsp;\u0026plusmn;\u0026thinsp;0.14\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eRepeated-measures ANOVA revealed significant changes in hip (p\u0026thinsp;=\u0026thinsp;0.003) and knee (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) flexion-extension RoM (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e). For the hip, post-hoc comparisons showed larger RoM at the highest assistance level: significance was met between F and A3 (p\u0026thinsp;=\u0026thinsp;0.024), T and A3 (p\u0026thinsp;=\u0026thinsp;0.024), and between the two assistance levels (p\u0026thinsp;=\u0026thinsp;0.016). Similarly, for the knee, higher RoM was registered in A3 modality compared to the other conditions (p\u0026thinsp;=\u0026thinsp;0.016 for F, p\u0026thinsp;=\u0026thinsp;0.002 for T, p\u0026thinsp;=\u0026thinsp;0.014 for A1). Additionally, a significant difference was also found between the T and the A1 modalities, with the latter reaching higher values. Ankle RoM did not reveal significant changes, suggesting that the interaction with the exoskeleton did not alter the angular trajectories of the distal joints.\u003c/p\u003e\n\u003cp\u003eSkewness resulted significantly altered only for the hip flexion-extension angle (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Specifically, no significant changes were observed between F and T modalities. However, when assistance was provided, both at A1 (p\u0026thinsp;=\u0026thinsp;0.019 for F, p\u0026thinsp;=\u0026thinsp;0.026 for T) and at A3 (p\u0026thinsp;=\u0026thinsp;0.003 for both F and T), skewness significantly increased (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e). This shift brought the values closer to zero, indicating that the joint curve became more symmetrical compared to the baseline condition, where the distribution was negatively skewed.\u003c/p\u003e"},{"header":"4. Discussion","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e4.1. Summary of the main results\u003c/h2\u003e \u003cp\u003eThe goal of this study was to evaluate the interaction between the Hypershell-X hip exoskeleton and a group of healthy participants during overground walking, with a focus on potential deviations from physiological gait. To this end, 16 participants performed four walking sessions under different levels of exoskeleton assistance, and joint kinematics were recorded and analysed. Our results showed no significant changes in spatiotemporal parameters and high correlation in the joint angular patterns across conditions, even if the highest assistance level led to significant alterations in hip and knee RoM and in hip skewness.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e4.2. Effects of minimal assistance on gait kinematics\u003c/h2\u003e \u003cp\u003eWhen comparing F and T conditions, no significant differences emerged in any of the spatiotemporal or kinematic parameters analysed, suggesting that the device did not interfere with physiological gait when no assistance was provided. This is an encouraging result, as it indicates that the overall structure and weight of the device do not hinder natural walking, which is an essential prerequisite for ensuring that any assistance delivered can be effectively integrated into the user\u0026rsquo;s movement (Camardella et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Moscatelli et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Proietti et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Stramel \u0026amp; Agrawal, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Therefore, we can conclude that the device\u0026rsquo;s transparency is high when only wearing Hypershell.\u003c/p\u003e \u003cp\u003eEven when low-level assistance (A1) was introduced, both spatiotemporal parameters and joint RoMs remained comparable to the free condition, supporting the hypothesis that the A1 modality does not alter physiological gait or the natural amplitude of joint motion. This aligns with the intended purpose of A1, which is designed for daily-life walking and therefore aims to provide minimal, subtle assistance. In addition, the angular profiles remained highly correlated across conditions (r\u0026thinsp;\u0026gt;\u0026thinsp;0.9), confirming that the overall shape and timing of the joint trajectories were mostly preserved.\u003c/p\u003e \u003cp\u003eInterestingly, skewness revealed a subtler modification of hip kinematics. In physiological gait, the hip flexion-extension angle naturally exhibits a negative skewness, reflecting that the joint spends a longer portion of the gait cycle in an extended configuration rather than in a flexed one. This pattern was maintained when assistance was not provided in the Transparent condition. In the A1 assistance configuration, however, skewness became significantly less negative, indicating a shift toward a more symmetric temporal distribution. This suggests that participants spent more time in a flexed configuration compared to the free condition. Such behaviour is consistent with the assistance strategy of the device: although the RoM did not change, the exoskeleton likely supported hip flexion enough to \u0026ldquo;shift\u0026rdquo; the angle profile upward (i.e., the hip was maintained in slightly greater flexion without altering the overall curve), thus increasing the time spent in flexion and making the distribution more symmetric.\u003c/p\u003e \u003cp\u003eIn summary, low-level assistance (A1) preserves the global gait pattern, as reflected by the unchanged RoM, stable spatiotemporal parameters, and high correlation of joint trajectories. This is consistent with previous findings in the literature: in fact, several studies reported no significant alterations in gait kinematics across different assistance levels provided with other devices for gait assistance, even in more complex scenarios such as when using the Lokomat, a lower-limb exoskeleton commonly employed for rehabilitation (Cherni et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Di Tommaso et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). In our case, the only detectable adaptation was a redistribution of the hip motion phases, possibly reflecting a mild adjustment to the assistance profile without affecting overall kinematic structure. The knee and ankle joints, instead, were not influenced by this assistance level, consistent with the fact that the exoskeleton directly acts on the hip, and further supporting the minimal impact of A1 on physiological gait.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e4.3. Effects of high assistance on gait kinematics\u003c/h2\u003e \u003cp\u003eUnlike what we observed for the A1 assistance level, which was recommended for low-intensity activities such as ground-level walking, significant alterations emerged when the exoskeleton provided a higher amount of assistance. The most evident change was the increase in hip RoM, driven by excessive flexion during the swing phase induced by the stronger flexion torque delivered by the device. Skewness was also affected, showing a shift toward less negative values, and therefore a more symmetric distribution of joint activation. In this case, the effect was much more pronounced than in the A1 condition: the higher assistance level generated a greater flexion peak during swing, combined with a reduced extension in late stance, resulting in an overall increase in the time spent in flexion.\u003c/p\u003e \u003cp\u003eInterestingly, the knee was also influenced by the high assistance level. The increased hip flexion lifted the limb more prominently during swing, leading to a significant increase in knee RoM compared to all other conditions. However, this modification was limited to the RoM, as knee skewness remained unchanged: knee flexion-extension angle is naturally characterized by a positive skewness, and the increased RoM at the knee did not substantially modify the temporal distribution of the curve. The ankle joint, instead, did not exhibit significant alterations, confirming that the effects of excessive assistance remain localized to the proximal joints, where the torque is directly applied.\u003c/p\u003e \u003cp\u003eDespite these joint-specific changes, neither the overall joint angular shape nor the spatiotemporal parameters were altered. The fundamental characteristics of gait were therefore preserved, even in high assistance trials. This suggests that when the device delivers an excessive amount of torque, the global gait pattern remains stable, while intrinsic features of the joints, particularly at the hip, where the exoskeleton acts directly, are modified. As a result, gait becomes less natural, although from a biomechanical perspective, the system still operates effectively. This observation also suggests that such a high assistance level might be more suitable for more demanding tasks, such as running or climbing, where a stronger flexion contribution is expected. In this scenario, no major kinematic alterations are expected, but rather a form of assistance that is coherent with the increased mechanical and energetic requirements of the task.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e4.4. Limitations and future works\u003c/h2\u003e \u003cp\u003eThis study presented a complete kinematic analysis of human gait during different levels of interaction with a powered hip exoskeleton. Although the results obtained indicate that the interaction with the exoskeleton led to minimal kinematic changes, this work did not directly investigate the underlying neural and motor control adaptation potentially induced by the exoskeleton. In particular, valuable information could be gained by analysing individual muscle activation profiles, which are commonly employed to study human-robot interaction (Moscatelli et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), as well as muscle coordination patterns, i.e., muscle synergies, obtained from EMG data (d\u0026rsquo;Avella et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Ivanenko et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). Such an approach would allow a deeper understanding of how the central nervous system adapts to the assistance and could provide stronger evidence supporting our results.\u003c/p\u003e \u003cp\u003eAnother limitation is that the study examined only a single, low-demanding walking task performed at a self-selected speed. As a result, the behaviour of the device in more challenging conditions, such as running, climbing stairs, or other tasks requiring greater mechanical effort, remains unexplored. Future studies should therefore assess the effects of the exoskeleton in these higher-demanding scenarios to better characterise and understand its assistance capabilities.\u003c/p\u003e \u003c/div\u003e"},{"header":"5. Conclusions","content":"\u003cp\u003eIn this study, we investigated the interaction between the Hypershell-X exoskeleton and a cohort of healthy subjects during overground walking through a comprehensive kinematic analysis. Overall, the device demonstrated a good level of preservation of the physiological gait, as both spatiotemporal parameters and joint kinematics were maintained across conditions. Only at the highest assistance level tested did we observe significant alterations in hip RoM and skewness, but the angular patterns remained strongly correlated with physiological gait. These findings suggest that the assistance delivered by the exoskeleton is well integrated into the user\u0026rsquo;s natural movement, although excessive support may lead to localized modifications at the hip. Future studies should focus on fully characterising the underlying motor control strategies during the interaction with the device by combining kinematic data with EMG-based analyses, such as muscle synergies. Moreover, evaluating the device in more demanding tasks will help explore its full assistance potential.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNM carried out experiments, analysed data, wrote and reviewed the paper. VL carried out experiments, analysed data, wrote and reviewed the paper. CB carried out experiments, analysed data, wrote and reviewed the paper. LMT reviewed the paper and was responsible for the supervision and acquisition of the funding. AS carried out experiments, analysed the data, wrote and reviewed the paper, participated in the conceptualization of the study and supervised the work. All authors have read and agreed to the published version of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFundings\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work is supported by the Italian Ministry of University and Research, under the complementary actions to the Plan of National Recovery and Resilience (PNRR)\u0026nbsp;\u0026ldquo;Fit4MedRob -Fit for Medical Robotics\u0026rdquo;\u0026nbsp;Grant (PNC0000007).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe dataset obtained during the study is available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthical statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was reviewed and approved by the Ethics Committee of the National Research Council (approval number N\u0026deg;231878/2025, approved on 24/06/2025), according to the Declaration of Helsinki.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBogue, R. (2015). Robotic Exoskeletons: A Review of Recent Progress. \u003cem\u003eIndustrial Robot\u003c/em\u003e, \u003cem\u003e42\u003c/em\u003e(1), 5\u0026ndash;10. https://doi.org/10.1108/IR-08-2014-0379\u003c/li\u003e\n\u003cli\u003eBryan, G. M., Franks, P. W., Song, S., Voloshina, A. S., Reyes, R., O\u0026rsquo;Donovan, M. P., Gregorczyk, K. N., \u0026amp; Collins, S. H. (2021). Optimized hip-knee-ankle exoskeleton assistance at a range of walking speeds. \u003cem\u003eJournal of Neuroengineering and Rehabilitation\u003c/em\u003e, \u003cem\u003e18\u003c/em\u003e(1), 152. https://doi.org/10.1186/s12984-021-00943-y\u003c/li\u003e\n\u003cli\u003eCalabr\u0026ograve;, R. S., Sorrentino, G., Cassio, A., Mazzoli, D., Andrenelli, E., Bizzarini, E., Campanini, I., Carmignano, S. M., Cerulli, S., Chisari, C., Colombo, V., Dalise, S., Fundar\u0026ograve;, C., Gazzotti, V., Mazzoleni, D., Mazzucchelli, M., Melegari, C., Merlo, A., Stampacchia, G., \u0026hellip; Italian Consensus Conference on Robotics in Neurorehabilitation (CICERONE). (2021). Robotic-assisted gait rehabilitation following stroke: A systematic review of current guidelines and practical clinical recommendations. \u003cem\u003eEuropean Journal of Physical and Rehabilitation Medicine\u003c/em\u003e, \u003cem\u003e57\u003c/em\u003e(3), 460\u0026ndash;471. https://doi.org/10.23736/S1973-9087.21.06887-8\u003c/li\u003e\n\u003cli\u003eCalafiore, D., Negrini, F., Tottoli, N., Ferraro, F., Ozyemisci-Taskiran, O., \u0026amp; de Sire, A. (2022). Efficacy of robotic exoskeleton for gait rehabilitation in patients with subacute stroke: A systematic review. \u003cem\u003eEuropean Journal of Physical and Rehabilitation Medicine\u003c/em\u003e, \u003cem\u003e58\u003c/em\u003e(1), 1\u0026ndash;8. https://doi.org/10.23736/S1973-9087.21.06846-5\u003c/li\u003e\n\u003cli\u003eCamardella, C., Porcini, F., Filippeschi, A., Marcheschi, S., Solazzi, M., \u0026amp; Frisoli, A. (2021). Gait Phases Blended Control for Enhancing Transparency on Lower-Limb Exoskeletons. \u003cem\u003eIEEE Robotics and Automation Letters\u003c/em\u003e, \u003cem\u003e6\u003c/em\u003e(3), 5453\u0026ndash;5460. https://doi.org/10.1109/LRA.2021.3075368\u003c/li\u003e\n\u003cli\u003eCardoso, A., Ribeiro, A., Carneiro, P., \u0026amp; Colim, A. (2024). Evaluating Exoskeletons for WMSD Prevention: A Systematic Review of Applications and Ergonomic Approach in Occupational Settings. \u003cem\u003eInternational Journal of Environmental Research and Public Health\u003c/em\u003e, \u003cem\u003e21\u003c/em\u003e(12), Article 12. https://doi.org/10.3390/ijerph21121695\u003c/li\u003e\n\u003cli\u003eCarpinella, I., Cattaneo, D., Abuarqub, S., \u0026amp; Ferrarin, M. (2009). Robot-based rehabilitation of the upper limbs in multiple sclerosis: Feasibility and preliminary results. \u003cem\u003eJournal of Rehabilitation Medicine\u003c/em\u003e, \u003cem\u003e41\u003c/em\u003e(12), 966\u0026ndash;970. https://doi.org/10.2340/16501977-0401\u003c/li\u003e\n\u003cli\u003eChen, W., Wu, S., Zhou, T., \u0026amp; Xiong, C. (2018). On the biological mechanics and energetics of the hip joint muscle-tendon system assisted by passive hip exoskeleton. \u003cem\u003eBioinspiration \u0026amp; Biomimetics\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(1), 016012. https://doi.org/10.1088/1748-3190/aaeefd\u003c/li\u003e\n\u003cli\u003eCherni, Y., Blache, Y., Begon, M., Ballaz, L., \u0026amp; Dal Maso, F. (2023). Effect of Robotic-Assisted Gait at Different Levels of Guidance and Body Weight Support on Lower Limb Joint Kinematics and Coordination. \u003cem\u003eSensors (Basel, Switzerland)\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e(21), 8800. https://doi.org/10.3390/s23218800\u003c/li\u003e\n\u003cli\u003ed\u0026rsquo;Avella, A., Portone, A., Fernandez, L., \u0026amp; Lacquaniti, F. (2006). Control of fast-reaching movements by muscle synergy combinations. \u003cem\u003eJournal of Neuroscience\u003c/em\u003e, \u003cem\u003e26\u003c/em\u003e(30), 7791\u0026ndash;7810. https://doi.org/10.1523/JNEUROSCI.0830-06.2006\u003c/li\u003e\n\u003cli\u003ede Looze, M. P., Bosch, T., Krause, F., Stadler, K. S., \u0026amp; O\u0026rsquo;Sullivan, L. W. (2016). Exoskeletons for Industrial Application and Their Potential Effects on Physical Work Load. \u003cem\u003eErgonomics\u003c/em\u003e, \u003cem\u003e59\u003c/em\u003e(5), 671\u0026ndash;681. https://doi.org/10.1080/00140139.2015.1081988\u003c/li\u003e\n\u003cli\u003eDi Tommaso, F., Tamburella, F., Lorusso, M., Gastaldi, L., Molinari, M., \u0026amp; Tagliamonte, N. L. (2023). Biomechanics of Exoskeleton-Assisted Treadmill Walking. \u003cem\u003e2023 International Conference on Rehabilitation Robotics (ICORR)\u003c/em\u003e, 1\u0026ndash;6. https://doi.org/10.1109/ICORR58425.2023.10304685\u003c/li\u003e\n\u003cli\u003eDing, Y., Kim, M., Kuindersma, S., \u0026amp; Walsh, C. J. (2018). Human-in-the-loop optimization of hip assistance with a soft exosuit during walking. \u003cem\u003eScience Robotics\u003c/em\u003e, \u003cem\u003e3\u003c/em\u003e(15), eaar5438. https://doi.org/10.1126/scirobotics.aar5438\u003c/li\u003e\n\u003cli\u003eFraiwan, L., \u0026amp; Hassanin, O. (2021). Computer-aided identification of degenerative neuromuscular diseases based on gait dynamics and ensemble decision tree classifiers. \u003cem\u003ePLoS ONE\u003c/em\u003e, \u003cem\u003e16\u003c/em\u003e(6), e0252380. https://doi.org/10.1371/journal.pone.0252380\u003c/li\u003e\n\u003cli\u003eFritz, H., Patzer ,Diane, \u0026amp; and Galen, S. S. (2019). Robotic exoskeletons for reengaging in everyday activities: Promises, pitfalls, and opportunities. \u003cem\u003eDisability and Rehabilitation\u003c/em\u003e, \u003cem\u003e41\u003c/em\u003e(5), 560\u0026ndash;563. https://doi.org/10.1080/09638288.2017.1398786\u003c/li\u003e\n\u003cli\u003eGassert, R., \u0026amp; Dietz, V. (2018). Rehabilitation robots for the treatment of sensorimotor deficits: A neurophysiological perspective. \u003cem\u003eJournal of NeuroEngineering and Rehabilitation\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e, 46. https://doi.org/10.1186/s12984-018-0383-x\u003c/li\u003e\n\u003cli\u003eGrimmer, M., \u0026amp; Zhao, G. (2024). Hip Exoskeleton for Cycling Assistance. \u003cem\u003eBioengineering\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(7), 683. https://doi.org/10.3390/bioengineering11070683\u003c/li\u003e\n\u003cli\u003eHsu, T.-H., Tsai, C.-L., Chi, J.-Y., Hsu, C.-Y., \u0026amp; Lin, Y.-N. (2023). Effect of wearable exoskeleton on post-stroke gait: A systematic review and meta-analysis. \u003cem\u003eAnnals of Physical and Rehabilitation Medicine\u003c/em\u003e, \u003cem\u003e66\u003c/em\u003e(1), 101674. https://doi.org/10.1016/j.rehab.2022.101674\u003c/li\u003e\n\u003cli\u003eIshmael, M. K., Gunnell, A., Pruyn, K., Creveling, S., Hunt, G., Hood, S., Archangeli, D., Foreman, K. B., \u0026amp; Lenzi, T. (2023). Powered Hip Exoskeleton Reduces Residual Hip Effort Without Affecting Kinematics and Balance in Individuals With Above-Knee Amputations During Walking. \u003cem\u003eIEEE Transactions on Bio-Medical Engineering\u003c/em\u003e, \u003cem\u003e70\u003c/em\u003e(4), 1162\u0026ndash;1171. https://doi.org/10.1109/TBME.2022.3211842\u003c/li\u003e\n\u003cli\u003eIvanenko, Y. P., Poppele, R. E., \u0026amp; Lacquaniti, F. (2004). Five basic muscle activation patterns account for muscle activity during human locomotion. \u003cem\u003eThe Journal of Physiology\u003c/em\u003e, \u003cem\u003e556\u003c/em\u003e(Pt 1), 267\u0026ndash;282. https://doi.org/10.1113/jphysiol.2003.057174\u003c/li\u003e\n\u003cli\u003eLee, Y., Roh, S., Lee, M., Choi, B., Lee, J., Kim, J., Choi, H., Shim, Y., \u0026amp; Kim, Y.-J. (2017). A flexible exoskeleton for hip assistance. \u003cem\u003e2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)\u003c/em\u003e, 1058\u0026ndash;1063. https://doi.org/10.1109/IROS.2017.8202275\u003c/li\u003e\n\u003cli\u003eLim, B., Lee, J., Jang, J., Kim, K., Park, Y. J., Seo, K., \u0026amp; Shim, Y. (2019). Delayed Output Feedback Control for Gait Assistance With a Robotic Hip Exoskeleton. \u003cem\u003eTrans. Rob.\u003c/em\u003e, \u003cem\u003e35\u003c/em\u003e(4), 1055\u0026ndash;1062. https://doi.org/10.1109/TRO.2019.2913318\u003c/li\u003e\n\u003cli\u003eMarimon, X., Mengual, I., L\u0026oacute;pez-de-Celis, C., Portela, A., Rodr\u0026iacute;guez-Sanz, J., Herr\u0026aacute;ez, I. A., \u0026amp; P\u0026eacute;rez-Bellmunt, A. (2024). Kinematic Analysis of Human Gait in Healthy Young Adults Using IMU Sensors: Exploring Relevant Machine Learning Features for Clinical Applications. \u003cem\u003eBioengineering (Basel, Switzerland)\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(2), 105. https://doi.org/10.3390/bioengineering11020105\u003c/li\u003e\n\u003cli\u003eMoscatelli, N., Brambilla, C., Lanzani, V., Molinari Tosatti, L., \u0026amp; Scano, A. (2025). Assessing Transparency of Robots, Exoskeletons, and Assistive Devices: A Systematic Review. \u003cem\u003eSensors\u003c/em\u003e, \u003cem\u003e25\u003c/em\u003e(14), 4444. https://doi.org/10.3390/s25144444\u003c/li\u003e\n\u003cli\u003eNepomuceno, P., Souza, W. H., Pakosh, M., Musselman, K. E., \u0026amp; Craven, B. C. (2024). Exoskeleton-based exercises for overground gait and balance rehabilitation in spinal cord injury: A systematic review of dose and dosage parameters. \u003cem\u003eJournal of Neuroengineering and Rehabilitation\u003c/em\u003e, \u003cem\u003e21\u003c/em\u003e(1), 73. https://doi.org/10.1186/s12984-024-01365-2\u003c/li\u003e\n\u003cli\u003eNurse, C. A., Wolf, D. N., Rodzak, K. M., Teater, R. H., Ice, C. C., Fine, S. J., Holtzman, E. C., \u0026amp; Zelik, K. E. (2025). Evaluating the Biomechanical Effects and Real-World Usability of a Novel Ankle Exo for Runners. \u003cem\u003eJournal of Biomechanical Engineering\u003c/em\u003e, \u003cem\u003e147\u003c/em\u003e(3), 031004. https://doi.org/10.1115/1.4067579\u003c/li\u003e\n\u003cli\u003ePark, Y.-H., Lee, D.-H., \u0026amp; Lee, J.-H. (2024). A Comprehensive Review: Robot-Assisted Treatments for Gait Rehabilitation in Stroke Patients. \u003cem\u003eMedicina (Kaunas, Lithuania)\u003c/em\u003e, \u003cem\u003e60\u003c/em\u003e(4), 620. https://doi.org/10.3390/medicina60040620\u003c/li\u003e\n\u003cli\u003eProietti, T., Crocher, V., Roby-Brami, A., \u0026amp; Jarrasse, N. (2016). Upper-Limb Robotic Exoskeletons for Neurorehabilitation: A Review on Control Strategies. \u003cem\u003eIEEE Reviews in Biomedical Engineering\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e, 1\u0026ndash;1. https://doi.org/10.1109/RBME.2016.2552201\u003c/li\u003e\n\u003cli\u003eSanz-Mor\u0026egrave;re, C. B., Martini, E., Meoni, B., Arnetoli, G., Giffone, A., Doronzio, S., Fanciullacci, C., Parri, A., Conti, R., Giovacchini, F., Fri\u0026eth;riksson, \u0026THORN;., Romo, D., Crea, S., Molino-Lova, R., \u0026amp; Vitiello, N. (2021). Robot-mediated overground gait training for transfemoral amputees with a powered bilateral hip orthosis: A pilot study. \u003cem\u003eJournal of NeuroEngineering and Rehabilitation\u003c/em\u003e, \u003cem\u003e18\u003c/em\u003e(1), 111. https://doi.org/10.1186/s12984-021-00902-7\u003c/li\u003e\n\u003cli\u003eSawicki, G. S., Beck, O. N., Kang, I., \u0026amp; Young, A. J. (2020). The exoskeleton expansion: Improving walking and running economy. \u003cem\u003eJournal of Neuroengineering and Rehabilitation\u003c/em\u003e, \u003cem\u003e17\u003c/em\u003e(1), 25. https://doi.org/10.1186/s12984-020-00663-9\u003c/li\u003e\n\u003cli\u003eSeo, K., Lee, J., Lee, Y., Ha, T., \u0026amp; Shim, Y. (2016). Fully autonomous hip exoskeleton saves metabolic cost of walking. \u003cem\u003e2016 IEEE International Conference on Robotics and Automation (ICRA)\u003c/em\u003e, 4628\u0026ndash;4635. https://doi.org/10.1109/ICRA.2016.7487663\u003c/li\u003e\n\u003cli\u003eSrivatsa, A. V., Aghamohammadi, N. R., Ryali, P., Scarpellini, A., Kim, J. J. H., Aggarwal, A., Wright, Z., Huang, F. C., Reinkensmeyer, D. J., Esmailbeigi, H., \u0026amp; Patton, J. L. (2025). Distribution Analysis for Diagnostics and Therapeutics of Motor Actions. \u003cem\u003eIEEE Journal of Biomedical and Health Informatics\u003c/em\u003e, \u003cem\u003ePP\u003c/em\u003e. https://doi.org/10.1109/JBHI.2025.3636009\u003c/li\u003e\n\u003cli\u003eStramel, D., \u0026amp; Agrawal, S. (2022). Assessing Changes in Human Gait with a Mobile Tethered Pelvic Assist Device (mTPAD) in Transparent Mode with Hand Holding Conditions. \u003cem\u003eProceedings of the IEEE International Conference on Biomedical Robotics and Biomechatronics (BioRob)\u003c/em\u003e, 1\u0026ndash;6. https://doi.org/10.1109/BioRob52689.2022.9925359\u003c/li\u003e\n\u003cli\u003eYandell, M. B., Tacca, J. R., \u0026amp; Zelik, K. E. (2019). Design of a Low Profile, Unpowered Ankle Exoskeleton That Fits Under Clothes: Overcoming Practical Barriers to Widespread Societal Adoption. \u003cem\u003eIEEE Transactions on Neural Systems and Rehabilitation Engineering\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e(4), 712\u0026ndash;723. https://doi.org/10.1109/TNSRE.2019.2904924\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Institute of Intelligent Industrial Technologies and Systems for Advanced Manufacturing (STIIMA), Italian National Research Council ","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Exoskeletons, Human-Robot Interaction, Gait, Assistive Devices, Wearable Robotics, Lower-limb kinematics","lastPublishedDoi":"10.21203/rs.3.rs-8367199/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8367199/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eWearable robotic devices such as lower-limb exoskeletons have been recently proposed to support mobility, assist activities of daily living, and even aid rehabilitation. Understanding and quantifying the quality of interaction between human and robot is essential to ensure physiological and effective support. In this study, we evaluated a transparent (i.e., non-assisting) modality and two levels of assistance (low and high) of the Hypershell-X hip exoskeleton during overground walking in a cohort of 16 healthy individuals. Our goal was to characterize the impact of the assistance on physiological gait from a kinematic perspective. The results suggest that the exoskeleton preserved natural gait characteristics across conditions: no significant changes were found in spatiotemporal parameters, and joint profiles on the sagittal plane remained highly correlated (r\u0026thinsp;\u0026gt;\u0026thinsp;0.9). Significant alterations emerged at the highest assistance level, which mainly affected the hip joint, where the assistive torque is directly applied. However, even in this experimental condition, the overall gait biomechanics was preserved. These findings indicate that the Hypershell-X provides functional and effective assistance without relevant alteration of the physiological structure of gait.\u003c/p\u003e","manuscriptTitle":"Impact of Hip Exoskeleton Assistance on Human Gait: A Kinematic Study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-16 13:39:40","doi":"10.21203/rs.3.rs-8367199/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":"76125417-27da-47cc-bc07-23b777ba3fbd","owner":[],"postedDate":"December 16th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-12-16T13:39:40+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-16 13:39:40","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8367199","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8367199","identity":"rs-8367199","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}