Mitigating Compensatory Movements in Prosthesis Users via Adaptive Collaborative Robotics
The objective of this paper is to address the issue of compensatory movements in prosthesis users by employing a novel human-robot collaboration framework. While prosthetic devices enable partial restoration of limb functionality, they often fail to fully replicate the natural mobility of limbs, leading to compensatory movements that result in discomfort, inefficiency, and physical strain over time. The proposed framework focuses on integrating a personalised mobility model into a constrained optimisation setting, guiding collaborative robots to reconfigure task environments. This promotes optimal user postures, balancing functionality and comfort during task performance.
Framework Overview
The framework includes a personalised mobility model that quantifies joint-specific functional limitations and the cost associated with compensatory movements. This cost, denoted $\Psi$, is measured by the deviation of functioning joints from a comfortable, neutral posture, which traditional methods in the literature often overlook. The constrained optimisation framework computes optimal user postures, ensuring task constraints while minimizing the weighted sum of deviations from both impaired and functional joints. Task constraints include maintaining safety distances and adhering to task objectives such as object positioning in the task space.
Experimental Evaluation
The validation of the approach involved a novel body-powered prosthetic device tailored for single-finger amputees. This prototype, enhancing grasping capabilities through synergistic closure with the hand, simultaneously imposes wrist constraints as part of the design considerations. In the experimental setup, a collaborative robot was positioned to assist prosthesis users wearing the finger prosthetic in handover tasks with varying lateral distances. The significance of integrating a user-specific mobility model was illustrated by comparing robotic assistance to traditional human cooperation in object handover scenarios.
Results and Implications
Empirical results demonstrated that embedding a user-specific mobility model in robotic assistance resulted in significant improvements in grasp efficiency and reduction of compensatory movements. Notably, robotic assistance achieved a noteworthy reduction in the cost of compensatory motions and substantially improved handover task duration by allowing users to adopt optimised postures that human partners often failed to facilitate. These findings suggest that collaborative robots could serve effectively as workplace and caregiving assistants, promoting better integration of prosthetic devices into daily tasks.
Future Directions
The study sparked potential avenues for further exploration, such as extending the framework to accommodate more complex prosthetic systems involving multiple joint impairments, including those affecting the elbow. Additionally, the development of intuitive user feedback mechanisms, such as visual or haptic interfaces, to guide users in configuring joint motions could be advantageous. Further research could explore the application of this framework in diverse task settings, enhancing overall usability and efficacy across various real-world scenarios involving prosthesis users.
In conclusion, the paper provides a robust foundation for addressing compensatory movements in prosthesis users. By leveraging adaptive human-robot collaboration, the framework optimizes task performance while enhancing user comfort, potentially transforming prosthetic device integration into everyday activities. Collaborative robots, equipped with user-specific mobility models and ergonomic task configurations, can significantly contribute to the development of inclusive and efficient human-prosthesis interactions.