Motion control of a four-wheel-independent-drive electric vehicle by motor imagery EEG based BCI system

This paper proposed a control structure of Four-Wheel-Independent-Drive electric vehicles by motor imagery Electroencephalography (EEG) based BCI system. First work is the BCI system, Emotiv EPOC+ is used to acquire the raw EEG data, Independent Component Analysis (ICA) is used to preprocess the motor imagery EEG, then Common Spatial Pattern (CSP) is used to extract the features which is most related to the ERD/ERS. Last but not least, Back Propagation Neural Network (BPNN) is used to classify the left and right, at the meantime, subject's threshold value of average band power is calculated to classify the accelerate and brake. In the test, there is about 84% accuracy rate of accelerate and brake and 89.92% for left and right classification. The second work is that we design and construct a real model FWID electric vehicle with environmental perception subsystem, including ultrasonic wave radars and cameras, which can avoid the risk when BCI command is wrong. We also construct driving motor imagery sample data collector to get subjects' motor imagery data for classifier to improve the accuracy of BCI system.