Research On Intelligent Upper Limb Prosthesis Training System With Bidirectional Neural Channel

The use of EMG prosthesis is inseparable from the training process.The traditional way is to use physical prostheses for training,which is inefficient,long and boring,and it is easy for patients to lose their motivation and confidence in training.The virtual training system has natural advantages in terms of intuitiveness,vividness,and interactivity,and has gradually become the main method of training.In order to improve the training efficiency of the patients and shorten the training time,this paper designed a smart upper limb prosthesis training system consisting of a multi-mode neural control module based on myoelectric signal and a sensory feedback module based on vibration stimulation,which has bidirectional nerves of motion and sensation Channel,and conducted a lively closed-loop interactive experiment.This article first analyzes the domestic and foreign research status of the artificial limb virtual training system,myoelectric control method and sensory feedback method,summarizes the shortcomings of the existing virtual training system,and determines the feedback method of the myoelectric control algorithm and sensory feedback module used by the system.Based on the functional requirements,the overall design was carried out.Using Visual C++ 6.0 and Open Inventor 3D graphics development library as a development platform,an intelligent upper limb prosthesis training system was established.Through the kinematic analysis of the arm-hand system,the upper limbs can be controlled.Design the virtual scene display window,collision detection module,automatic demonstration mode,and control training mode to realize the main functions of the system.Integrate multi-mode neural control and sensory feedback functions in the system.In the control training mode,display the classification result of the classifier and the action label of the action intention decoding through the interactive interface;use the spring-damping model as the contact model in the virtual environment,analyze the virtual force under different contact conditions,and design the vibration feedback strategy Generate control information of the vibration element to form a closed control loop.In order to verify the effectiveness of the virtual training system designed in this paper,a live closed-loop interactive experiment was conducted.Compared with the same training time,the traditional training and vital closed-loop interactive training have their respective success rates and action decoding time.The efficiency of the former reaction training and the proficiency of the latter reaction coding control prove the effectiveness of the virtual training system established in this paper in shortening the training time of patients and improving the ability of patients to operate prostheses.