Patients suffering from neurological and musculoskeletal disorders often experience impaired upper limb function, significantly reducing their quality of life. In recent years, wearable robots have emerged as a promising solution to facilitate rehabilitation and assist in daily activities. Among these, tendon-driven actuation has been widely adopted; however, such systems face challenges in achieving precise position control compared to direct motor-driven systems. This is primarily due to the hysteresis and backlash resulting from the high compliance and elasticity of tendons, necessitating effective compensation strategies. In this paper, we implement an embedded compact sensor system for end-effector-level position tracking in a soft wearable robot designed for forearm pronation/supination and grasping motions. By integrating sensors at the end-effector, we enable real-time motion data acquisition and establish a closed-loop feedback mechanism that effectively compensates for the limitations of tendon-driven actuation, thereby enhancing overall control accuracy. Based on the embedded end-effect-level sensing system, we introduce a novel wearable robot kit for motion tracking that comprising two parts: a sensor-only exosuit for real-time capture of user hand and forearm movements, and a motor-equipped exosuit that replicates and assists movements based on the sensor feedback. This Leader-Follower Control Mode allows for accurate capture and rapid response to user motion intent, offering a new solution for applications in tele-control, mirror therapy, and motion synchronization.

Human beings adapt their motor patterns in response to their surroundings, utilizing sensory modalities such as visual inputs. This context-informed adaptive motor behavior has increased interest in integrating computer vision algorithms into robotic assistive technologies, marking a shift towardscontext aware control. However, such integration has rarely been achieved so far, with current methods mostly relying on data-driven approaches. In this study, we introduce a novel control framework for a soft hip exosuit, employing instead a physics-informed computer vision method grounded on geometric modeling of the captured scene for assistance tuning during stairs and level walking. This approach promises to provide a viable solution that is more computationally efficient and does not depend on training examples. Evaluating the controller with six subjects on a path comprising level walking and stairs, we achieved an overall detection accuracy of$93.0\pm 1.1\%$. Computer vision-based assistance provided significantly greater metabolic benefits compared to non-vision-based assistance, with larger energy reductions relative to being unassisted during stair ascent ($-18.9 \pm 4.1\%$vs.$-5.2 \pm 4.1\%$) and descent ($-10.1 \pm 3.6\%$vs.$-4.7 \pm 4.8\%$). Such a result is a consequence of the adaptive nature of the device, enabled by the context aware controller, that allowed for more effective walking support: i.e. the assistive torque showed a significant increase while ascending stairs ($+33.9\pm 8.8\%$) and decrease while descending stairs ($-17.4\pm 6.0\%$) compared to a condition without assistance modulation enabled by vision. These results highlight the potential of the approach, promoting effective real-time embedded applications in assistive robotics.

Sitting, standing, and walking are fundamental activities crucial for maintaining independence in daily life. However, aging or lower limb injuries can impede these activities, posing obstacles to individuals' autonomy. In response to this challenge, we developed the LM-Ease (lower-limb movement ease), a compact and soft wearable robot designed to provide hip assistance. Its purpose is to aid users in carrying out essential daily activities such as sitting, standing, and walking. The LM-Ease features a fully actuated tendon-driven system that seamlessly transitions between assistance actuation profiles tailored for sitting, standing, and walking movements. This device provides the user with gravity support during stand-to-sit, and offers hip extension assistance pulling force during sit-to-stand and walking. Our preliminary results show that with the LM-Ease, healthy young adults (n$=$8) had significantly lower muscle activation: average reduction of 15.6% during stand-to-sit and 17.8% during sit-to-stand. Furthermore, with LM-Ease, participants demonstrated a 12.7% reduction in metabolic cost during ground walking. These evidences suggest that the LM-Ease holds potential in reducing muscular activation and energy expenditure during these fundamental daily activities. It could serve as a valuable tool for individuals seeking assistance in enhancing lower limb mobility, thereby bolstering their independence and overall quality of life.

The intention-detection strategy used to drive an exosuit is fundamental to evaluate the effectiveness and acceptability of the device. Yet, current literature on wearable soft robotics lacks evidence on the comparative performance of different control approaches for online intention-detection. In the present work, we compare two different and complementary controllers on a wearable robotic suit, previously formulated and tested by our group; a model-based myoelectric control (myoprocessor), which estimates the joint torque from the activation of target muscles, and a force control that estimates human torques using an inverse dynamics model (dynamic arm). We test them on a cohort of healthy participants performing tasks replicating functional activities of daily living involving a wide range of dynamic movements. Our results suggest that both controllers are robust and effective in detecting human–motor interaction, and show comparable performance for augmenting muscular activity. In particular, the biceps brachii activity was reduced by up to 74% under the assistance of thedynamic armand up to 47% under themyoprocessor, compared to a no-suit condition. However, themyoprocessoroutperformed thedynamic armin promptness and assistance during movements that involve high dynamics. The exosuit work normalized with respect to the overall work was$68.84 \pm 3.81\%$when it was ran by themyoprocessor, compared to$45.29 \pm 7.71\%$during thedynamic armcondition. The reliability and accuracy of motor intention detection strategies in wearable device is paramount for both the efficacy and acceptability of this technology. In this article, we offer a detailed analysis of the two most widely used control approaches, trying to highlight their intrinsic structural differences and to discuss their different and complementary performance.