Bioinspired Segmented Hybrid Bending Pneumatic Actuators for Rehabilitation/Assisting Training

Abstract

Soft pneumatic actuators/robotics have received dramatic attention in rehabilitation/assisting scenarios, owing to their intrinsic elasticity and simple control strategies for producing desired motions. However, most existing soft pneumatic bending actuators show homogenous/continuous deformation and the profile mismatch between the SPBAs and interacted objects, which hinder compatible motion and precise operations. This paper presents the design, modeling, and testing of pneumatic bioinspired segmented hybrid bending actuators (SHBAs) for rehabilitation interactions. The segmented structure inspired by the anatomy of the human finger, i.e., the soft-joint/rigid-bone configuration, makes the proposed actuator match the finger better than those of the traditional continuous structure. To efficiently predict the bending angles and the motion trajectories of the bioinspired actuator, a static model was developed relating the geometrical parameters. In addition, we build a finite-element model to quantitatively verify the experimental results and theoretical analysis simultaneously. Furthermore, underactuated finger rehabilitation gloves with SHBAs demonstrated their potential in assisting ADLs scenarios.