Multiple-Channel Real Time Filtering for a Myoelectric Prosthetic Hand-Arm Robot System

TL;DR

This paper introduces a multi-channel real-time filtering approach and a predictive control algorithm to enhance the stability and accuracy of a myoelectric prosthetic hand-arm system, demonstrating improved performance over previous methods.

Contribution

It proposes a novel multi-dimensional filtering method combined with a predictive control algorithm for real-time prosthetic control, improving signal stability and control accuracy.

Findings

Enhanced control signal stability and quality

Reduced processing time for real-time operation

Demonstrated improved performance in experiments

Abstract

On the base of the developed master-slave prosthetic hand-arm robot system, which is controlled mainly based on signals obtained from bending sensors fixed on the data glove, the first idea deduced was to develop and add a multi-dimensional filter into the original control system to make the control signals cleaner and more stable at real time. By going further, a second new idea was also proposed to predict new control information based on the combination of a new algorithm and prediction control theory. In order to fulfill the first idea properly, the possible methods to process data in real time, the different ways to produce Gaussian distributed random data, the way to combine the new algorithm with the previous complex program project, and the way to simplify and reduce the running time of the algorithm to maintain the high efficiency, the real time processing with multiple channels of the sensory system and the real-time performance of the control system were researched. Eventually, the experiment on the same provided robot system gives the results of the first idea and shows the improved performance of the filter comparing with the original control method.