A 3-D finger motion measurement system via soft strain sensors for hand rehabilitation

Abstract

Finger mobility plays a crucial role in everyday lives and is especially a leading indicator during hand rehabilitation and assistance tasks. In this study, a soft sensor-based three-dimensional (3-D) finger motion measurement system is proposed. Specifically, we report the use of a soft rope-embedded microfluidic sensor with high sensitivity and responsivity for strain sensing in the proposed system. These microfluidic strain sensors are proved to have the advantages of being easy to fabricate, chemically inert, with low hysteresis and good flexibility in their output signals. Besides, kinematic modeling of different finger joints are also provided to investigate the reasonable sensor locations for obtaining accurate measurements. Algorithms to identify the varied non-coplanar motions (abduction/adduction and flexion/extension joint angles) are presented by means of tracking the locations of the first metacarpal bone in multiple thumb postures. As a demonstration of their potential, the rope-embedded strain sensing units were used as curvature sensors mounted on the dorsum of target joints. The soft sensor-based 3-D finger motion measurement system and algorithms are experimentally verified by comparison with a camera-based motion capture device. Placing sensors at the optimal locations, joint angle measurement can reach up <3.5-deg accuracy. The proposed algorithms for improving the accuracy of the sensing system can be extended to other complicated finger joints.