Brain Computer Interface Based on Motor Imagery for Mechanical Arm Grasp Control

This paper puts forward a brain computer interface (BCI) system to realize the hand and wrist control using ABB Mechanical Arm. BCI gathers four kinds of motor imaginary (MI) tasks (hand grasp, spread, flexion extension) electroencephalogram (EEG) signals from 30 electrodes. It utilizes two fifth-order Butterworth Band-Pass Filter (BPF) with different bandwidths normalization method achieve raw MI EEG preprocessing. The main challenge feature extraction is extract enough representative features classify them. proposed extracts eleven in time domain time-frequency uses mutual information reduce large dimension extracted features. In addition, applies single convolutional layer Convolutional neural networks (CNN) filters implement quaternary classification tasks. Compared early researches, accuracy this increased by about 35%. actual mechanical arm grasping experiments verifies that has good adaptability.