Research And Experiment Of Intelligent Wheelchair Based On SSVEP EEG

In recent years,with the increasing aging of the population and the increasing number of disabled and handicapped people,solving their daily travel has become an urgent social problem.The rapid development of brain-computer interface technology and its increasingly widespread use make it an important means to solve the above phenomenon.It is of great practical significance to combine brain-computer interface with the wheelchair The research in the field of brain-computer interface has become a hot spot because of its high signal-to-noise ratio and no training.In order to explore the use of advanced brain-computer interface technology for helping elderly and disabled people to get around in daily life,this paper studies a high-performance EEG intelligent wheelchair system based on steady-state visual evoked potentials,uses a portable high-resolution device to develop the interface,optimizes feature extraction recognition,redesigns the control system for the wheelchair,and is able to achieve stable EEG control of the wheelchair through experiments.The research in this paper mainly includes the following aspects.(1)A visual stimulator was designed,which was implemented by using square wave modulation to encode frame section images and Psychtoolbox programming,followed by experiments using photosensitive sensors to verify the accuracy of the flicker frequency of the visual stimulator,and finally by collecting the EEG signal generated when the subject gazed at the visual stimulus for spectral analysis so as to obtain a valid steady-state visual evoked potential signal.(2)The feature extraction recognition of steady-state visual evoked potentials was optimized,and the results of several algorithms were analyzed by using self-picked datasets and Tsinghua University datasets,and finally a more efficient filter group-based typical correlation analysis algorithm was selected.For the individual variability of subjects,pre-screening optimal frequency was proposed,and EEG data were collected under the optimal frequency stimulation,and the final analysis of recognition was found to have improved accuracy and stability.(3)The EEG intelligent wheelchair system was built and completed.The system uses the SDK software package to develop the communication interface,which realizes the transfer of the EEG data collected online to MATLAB,intercepts the evoked signal through programming and performs pre-processing and feature extraction recognition,and finally can get the control commands for wheelchair.The wheelchair control system was designed according to the functional requirements,and the system established communication with the brain-machine interface part to realize the control command transmission,so as to control the wheelchair precisely.(4)An experimental overall evaluation of the EEG intelligent wheelchair system was designed.First,the optimal stimulation frequency of the subject was selected through a detection pre-experiment,and then a brain-controlled cart was used to experimentally identify and control a single command several times,and the experimental results showed that 95.94% recognition accuracy could be achieved,and the stability of command transmission and safe and reliable obstacle avoidance were realized,which verified the feasibility of the system.Finally,brain-controlled wheelchair experiments were conducted in real environments and under various maps,and the results showed that the subjects were able to complete the tasks efficiently with an accuracy rate of 95.25% or more,and its real-time and controllability could meet the design requirements.