Controlling a Rehabilitation Robot with Brain-Machine Interface: An approach based on Independent Component Analysis and Multiple Kernel Learning

Abstract

Patients suffering from severe motor disabilities usually require assistance from other people when doing rehabilitation exercises, which causes the rehabilitation process to be time-consuming and inconvenient. Therefore, we propose an automatic feature extraction method for a brain-machine interface that allows patients to control a robot using their own brain waves. A brain–machine interface (BMI) based on the P300 event-related potential (ERP), called Brain Controlled Rehabilitation System (BCRS), was developed to detect the intentions of patients. Using the BCRS, patients can communicate with the robot through their brain waves. However, deciding how to obtain an automatically extracted, useful EEG signal is a difficult and important problem for BMI research. In this paper, Independent Component Analysis – Multiple Kernel Learning (ICA-MKL) is used to directly extract a useful signal and build the classification mode for BCRS. The results reveal that this method is useful for automatically extracting the P300 signal and the accuracy is better than MKL. In additional, the same method can be extended into any motor imaginary area and the accuracy of ICA-MKL for brain imaginary data is also good to removing eye-blink artifacts and the accuracy performance is also good.