Design And Performance Comparative Analysis Modular Of Closed-loop Brain-machine Interfaca System

The brain-machine interface system can control the external devices by cstablishing the information transmission pathway between the brain and the external devices,and provides an effective method to restore the damaged motor function of human beings.It can effectively alleviate the contradiction between the supply and demand of medical rehabilitation resources,so it is of great significance to study the brain-machine interface system.The system mainly includes modules of brain,decoder,encoder and external equipment,etc.In order to improve the control accuracy of external equipment,an auxiliary controller module is introduced.On the basis of the improved single-joint information transmission,this paper studies the design and comparison of the decoder and the auxiliary controller module.The above research contents are based on the simulation implementation,and the specific contents are as follows:1.Information pathways in the brain into the artificial limb problem,this article is based on multivariate nonlinear fitting,support vector regression,k-nearest neighbor,artificial neural networks,Wiener filter,Kalman filter,respectively,design the decoder,the cerebral cortex area average discharge rate relevant to the movement into artificial prosthesis can accept joint torque information.The decoder training data and test data are obtained by the model.Finally,the test results show that the decoder based onlong short-term memory neural network has the best performance.2.In order to obtain a better tracking effect for the desired trajectory,an auxiliary controller is designed based on model-free control,simultaneous perturbation stochastic approximation,iterative feedback tuning algorithm and model predictive control,and constructs a closed-loop brain-machine interface system.The tracking effect of the four auxiliary controllers in tracking the desired trajectory and the time consumed in calculating the control output were tested and compared.The test shows that the better the tracking effect of the desired trajectory among the above-mentioned auxiliary controllers is,the longer the calculation time will be consumed.Therefore,in the process of building a closed-loop BMI system,the auxiliary controller should be reasonably selected according to the actual requirements.3.In order to verify the effect of the designed decoder and auxiliary controller module in practical application,this paper first uses the optimal decoder and auxiliary controller to con-struct a closed-loop brain-machine interface system.Secondly,an actual joint prosthesis is intro-duced into the system as an external device,and experiments are designed to verify the recovery effect of single joint movement.The experiments show that the closed-loop brain-machine in-terface system can accurately recover the actual motion of the prosthetic joint and make the end of the prosthetic joint reach the specified position,which proves the effectiveness of the decoder and auxiliary controller module designed in this paper.In this paper,the decoder and the auxiliary controller modules in the closed-loop brain-machine interface system are designed and the performance is compared and analyzed,which lays a foundation for the research of the whole system.In general,based on the designed decoder and auxiliary controller,the closed-loop brain-machine interface system constructed in this pa-per can accurately track the desired trajectory,and has a good effect on the recovery of single joint motion.The research results of this paper can further improve the research of closed-loop brain-machine interface system and provide a better basis for motion recovery.