Design of Mobile Robot Control System Based on EMG Signal

Abstract

In the current research, there are three main control methods for intelligent robot car, which are manual control, remote control and automatic control. And in the field of remote control, whether it is the direct use of remote control or the remote control keys are virtually installed in the mobile phone APP, there is a lack of direct interaction with human. To solve this problem, an intelligent robot car control system was developed to wirelessly control the intelligent car based on electromyography(EMG) signals in this paper. The intelligent robot car responds to the operator's different gestures by moving forward, backward, turning left and turning right. Firstly, EMG signals of different gestures of the forearm are collected, and the collected EMG signals are amplified, filtered and de-noised in MATLAB. Then, the data are processed by corresponding algorithms such as active segment detection, feature extraction and pattern classification on PC to obtain the recognition results of various gestures. Finally, different gestures are analyzed into corresponding commands and the signals are transmitted to the robot car through Bluetooth transmission so as to realize the real-time control of gestures to the robot car motion state. After several experiments, the intelligent car control system based on EMG can effectively control the robot car motion state according to the operator's gesture, and the recognition rate of the selected four gestures is more than 95%, the recognition rate is high, the system performance is good, and has good operability and stability.