Brain-vehicle Interactive Motion Control Based on Improved Queuing Network

Aiming at the blank in the existing brain- controlled vehicle driving, this paper establishes a brain-vehicle interactive motion control model based on improved queuing network, and studies the effects of brain-vehicle interactive motion control proficiency and brain-vehicle interaction related parameters on vehicle driving performance. The results show that with the increasing number of driving tasks, the overall task completion time and lateral deviation of brain-controlled vehicles are significantly reduced, and the out-of-bound rate and false-stop rate are also significantly decreased. It indicates that after a certain period of training and adaptation, the human can gradually master the related skills and control methods of brain-controlled vehicles, and that using electroencephalogram (EEG) signals to control vehicles will see safer and more efficient characteristics. When the control accuracy of the brain over the car is low (55% or so), the track of the car is chaotic at different response time. The longer the response time is, the more difficult for the car to follow the existing trajectory, and the more the steering wheel angle changes. It shows that the longer the interval set for the car to receive EEG signals, the weaker the brain’s control over the car, and the greater the offset of the car. As the proficiency of the subjects increases, and the interval of the vehicle receiving EEG signals becomes shorter, the success rate of the vehicle completing the whole driving task becomes greater.