Development And Application Of A Hybrid BCI System

Brain-computer interface(BCI) establishes a new communication channel between the human brain and the external devices. Current electroencephalogram(EEG) based BCI technologies mainly focus on how to independently use steady state visual evoked potential(SSVEP), motor imagery or other signals to recognize human intentions and generate several control commands. However, the generated control commands of these methods are limited and cannot control a robot to provide satisfied service to the user. Taking the advantages of SSVEP and motor imagery, this paper aims to design a hybrid BCI system that can provide multi BCI control commands to the robot. In this hybrid BCI system, three SSVEP signals are used to control the robot to move forward, turn left and turn right; one motor imagery signal is used to control the robot to execute the grasp motion; ? rhythm is used as a switch between the walking task and the grasping task.The whole study is divided into the offline experiment part and the online experiment part. In the offline experiment part, the electrode positions for collecting EEG signals were selected, the visual stimulator for generating SSVEPs was designed and the signal processing algorithms such as canonical correlation analysis(CCA) and short-time Fourier transform(STFT) for EEG classification were studied. In the online experiment part, the effect of the entire system was verified on a simulation platform(SIGVerse) and a real humanoid robot(NAO), respectively. The experimental results show that all of the three subjects are able to successfully use this hybrid BCI system with relative ease. The average accuracy of the walking task, grasping task and switching task of the simulated robot was 91.8%, 100% and 86.7%, respectively; and the average accuracy of the NAO robot control system was 89%, 93.3% and 86.7%, respectively.