Optimal design for individualised passive assistance

Abstract

Assistive devices are capable of restoring independence and function to people suffering from musculoskeletal impairments. Traditional assistive exoskeletons can be divided into active or passive devices depending on the method used to provide joint torques. The design of these devices often does not take into account the abilities of the individual leading to complex designs, joint misalignment and muscular atrophy due to over assistance at each joint. We present a novel framework for the design of passive assistive devices whereby the device provides the minimal amount of assistance required to maximise the space that they can reach. In doing so, we effectively remap their capable torque load over their workspace, exercising existing muscle while ensuring that key points in the workspace are reached. In this way we hope to reduce the risk of muscular atrophy while assisting with tasks. We implement two methods for finding the necessary passive device parameters, one looks at static loading conditions while the second simulates the system dynamics using level set methods. This allows us to determine the set of points that an individual can hold their arms stationary, the statically achievable workspace (SAW). We show the efficacy of these methods on a number of case studies which show that individuals with pronounced muscle weakness and asymmetric muscle weakness can have restored SAW restoring a range of motion.