A novel design and development on bioimpedance-based wheelchair control

Abstract

This article presents a novel design and development on bioimpedance-based wheelchair control for the disabled people and the elderly. We use three electrodes to measure two channels of bioimpedance from the trapezius muscle. Bioimpedance changes when there is a movement in the segment of trapezius muscle. We can classify six types of motions resulting in six operation capabilities for wheelchair control based on six types of shoulder movements, i.e. left shoulder up, right shoulder up, and both shoulder up for short time and long time. Our system is composed of the modified Howland current bridge circuit, which supplies the 0.5 mA ac current to the measurement system at the frequency of 50 kHz. NI PCI-6250 DAQ board was adopted to collect the data and Labview 8.2 was used to implement the signal processing and control system. Algorithms applied in the system are an automatic threshold value adjustment, which adapt its value to the measured signal. Pump value detection is used to detect the unexpected large change of the signal to avoid the wrong operation. Results indicate that the change of signal according to the shoulder movement is very stable. Moreover, we can use the shoulder movement to control LED on Labview8.2 with an accuracy of 100%.