Somatosensitive inverse artificial muscle for wearable robots

Abstract

Soft robotics have demonstrated significant potential in rehabilitation and wearable devices, attributed to their intrinsic compliance and safe interaction with humans. However, developing soft wearable robots with large deformation to assist the movement of human joints and simultaneously sensing the deformation remains challenging. This paper introduces a novel design of a type of compact somatosensitive inverse artificial muscle (SIAM) by directly coating sensing elements on the surface of soft actuators. We provided a theoretical model and conducted finite element simulations to analyze the impact of various parameters on the resistance of the proposed SIAM, and verified the model experimentally. Preliminary results show that the proposed SIAM is able to perform 200 % elongation with real-time length feedback. The maximum and minimum resistance of the SIAM changed by less than 5 % even after over 1000 working cycles. We further validated the effectiveness of the proposed SIAM on wearable robots in two working scenarios: elbow flexion rehabilitation and wrist flexion rehabilitation with the reduction of corresponding muscles’ activation levels from 100 % to 22.2 % and 24.3 %, respectively. Meanwhile, the SIAM could detect the joint angles with a maximum error of 2 degrees. This work provides an alternative design for soft wearable robots with high-fidelity and endurable sensing abilities.