NAO Robot Walking Control System Based on Motor Imagery

Abstract

Abstract Brain-computer interface (BCI) technology refers to the spontaneous imagination of a certain mode of movement in the brain to communicate with the computer or control external equipment. This technology helps the limb rehabilitation of paralysed and stroke patients. BCI system will fix electrodes on the experimenter’s brain. This paper designs a brain-computer interface system based on motor imagery, and uses it to control the movement of NAO robot. The system is mainly divided into three modules: signal acquisition, signal processing and robot control. In the aspect of signal acquisition, EEG cap, conductive paste, EEG amplifier and the designed upper computer software system are used to realize the design. Computer storage and processing of data. In signal processing, this paper designs a signal feature method based on Common Spatial Pattern (CSP) and Local Characteristic Scale Decomposition (LCD), and then classifies the features into commands to control NAO robots. In the aspect of communication control, the instruction coding is set in advance, and the classification result is set as the instruction to control the different actions of the robot. Socket communication is carried out by client-server mode using UDP protocol. The experimental results of the study also obtained a high accuracy.