Font Size: a A A

The Virtual Driving System Based On Brain-computer Interface

Posted on:2018-09-14Degree:MasterType:Thesis
Country:ChinaCandidate:X PanFull Text:PDF
GTID:2348330536478222Subject:Control engineering
Abstract/Summary:PDF Full Text Request
Brain-computer interfaces(BCI)and virtual reality(VR)both are popular fields of scientific research in current era.Brain-computer interfaces do not use the way which is human's physiological character of neuromuscular transmission,but the way to decode the pattern of brain activity,to realize communication and interaction between human and outside world.The brain-computer interfaces can give those severely handicapped the posibility to communicate and interact with the outside.But in some situations it is impossible for using outside hardware devices to avoid the expensive costs and the underlying unsafety.However,the appearing of virtual reality solves this problem.On the one hand,it can give those patients necessary devices for rehabilitation train and cut down the cost of using expensive hardware equipment.On the other hand,compared to realistic hardware equipment,the high verisimilitude of virtual reality can get the same or better result and doesn't take risks to patients.Moreover,those trains with the devices of virtual reality may have a better result than that with realistic devices.And it can be used as a game or a fun for users to enjoy life.At first,this article shows a novel P300 based BCI control of a simulated vehicle in a virtual city environments.This system has abandoned physical input devices which are used in most of the existing virtual reality platforms such as mouses,keyboards and handles.Users are able to use P300 signals to drive a vehicle and roam in a virtual and realistic city.Here,the self-adaptive Bayesian Linear Discriminant Analysis algorithm is used for the detection and classification of P300.And the displaying of P300 employed a 3D-P300 characters paradigm based on a 3d rendering engine of OSG.In the process of P300 training,all 8 participants` recognition accuracies reached more than 90% through cross validation.In the process of driving,P300 characters used in the process of virtual driving,as control commands of the virtual driving,can achieve functions such as “start”,“turn left ”,“turn right”,“keep”,“stop” and so on.After experiments,every subject need to evaluate their own states in the psychological state assessment,results of which shows that this system can pleasure users as a joyful game and is beneficial for subjects to be familiar with and understand brain-computer interface.Then,this article shows a virtual driving game controlled by the hybrid brain-computer interface.Different from the before-mentioned system,this system employs a hybrid BCI control mode of combination between P300 and motor imagery.It combines the advantage of multi-degree of freedom of P300 and the advantage of high information transfer rate of motor imagery.So it is more excellent than ordinary BCI in control mode.Besides,the driving game designed by the Unity game engine has high playability,vivid and realistic virtual reality scene.In the realization of driving control,the hybrid BCI system produces control instructions or control data,and does network transmission via the TCP/IP communication protocol.When the game system receives the control command or control data,the vehicle responses to those commands such as “acceleration”,“deceleration”,“turn left”,“turn right”,“keep” and “stop”.In this part,we designed two game scene to verify the reliability of the system,which are the game of “Hitting boxes” and the game of “ Surrounding the lake”.In the process of games,6 subjects finished the goal of game excellently without collisions.The result of experiment indicated this system has good playability and reliability.
Keywords/Search Tags:Brain-computer interface, Hybrid BCI, P300, Motor imagery, Virtual reality(VR), Bayesian Linear Discriminant Analysis
PDF Full Text Request
Related items