Research On Modeling And Control Of FES Rehabilitation Vehicle System

With the increase in the number of stroke/spinal cord injury patients in China,the number of patients with lower limb paralysis is also increasing year by year,and there is an urgent need for effective rehabilitation medical equipment to assist patients in rehabilitation training.The Functional Electrical Stimulation(FES)rehabilitation vehicle system stimulates the muscles of the lower extremities in corresponding patterns to promote coordinated alternate muscle contractions and active training for hemiplegic patients,which is conducive to the restoration of motor functions in paralyzed patients.However,the effect of FES rehabilitation vehicle training is closely related to the FES stimulation mode,and the different height,weight and other personal parameters of each patient as well as the vehicle model may lead to different FES stimulation modes.In addition,the FES rehabilitation vehicle open-loop control system has poor stability and safety problems.Therefore kinetic and Hill-Huxley mathematical models are established in this thesis to find the appropriate FES working mode for each individual,and a closed-loop control system based on constant speed training is designed for the FES rehabilitation vehicle system to make the patients more stable and safe during the riding training process.The main research of this thesis is as follows:(1)Establishing the stimulation mode of the FES rehabilitation vehicle system.Firstly,the kinematic and kinetic model of the riding motion of the FES rehabilitation vehicle is considered as sagittal motion in the plane,and the mathematical expressions of the moment and crank angle at the crank are derived based on the established model.Secondly,the lower limb free swing experiment and electrical stimulation experiment are designed to derive the mathematical expressions of the knee joint moment and crank angle based on the Hill-Huxley model.Finally,the mathematical models of crank moment and knee joint moment are substituted into the target efficiency function to solve the FES stimulation pattern(crank angle range)for each subject.(2)Establishing the FES rehabilitation vehicle riding speed model.Firstly,the data acquisition experiments are conducted on the data acquisition system platform of the FES rehabilitation vehicle,and the collected data are pre-processed by abnormal data rejection,limiting filtering and mean filtering methods,and the processed data are divided into training set and test set.Secondly,the data are collected using Genetic Algorithm-Least Square Support Vector Machine model is used to train the training set to obtain the FES rehabilitation vehicle riding speed model.Finally,the test set is fed into the trained model for validation,and the prediction results are obtained with a mean absolute error value MAE of 0.1769 and a coefficient of determination value R2 of 0.9818.This result indicates that the model has a high reliability in predicting the riding speed of the FES rehabilitation vehicle,so this model is selected as the controlled object in the controller simulation validation.(3)Research and analysis of the control method based on the FES rehabilitation vehicle system.Firstly,the basic principle of particle swarm optimization neural network PID(PSO-BP-PID)control algorithm is introduced.Secondly,the S function of PSO-BP-PID control algorithm and FES rehabilitation vehicle riding speed model is written,and the control simulation model is built on Simulink.Finally,the simulation verifies the performance and reliability of the controller designed by PID control algorithm and PSO-BP optimization PID The performance and response speed of the controller designed by the PID control algorithm and the PSO-BP optimized PID algorithm are verified in the final simulation.The simulation results show that the controller established by the PSO-BP optimized PID algorithm has smaller overshoot and shorter time to reach steady state.(4)FES rehabilitation vehicle control system construction and experimental verification.Firstly,the design scheme of FES rehabilitation control system is introduced.Secondly,the software system and hardware system of FES rehabilitation vehicle are built.The hardware system selects the hardware according to the design requirements and uses the communication methods of UART and UDP to realize the communication.Then,seven subjects with different physical characteristics are recruited to ride in three training modes:conventional training,fixed pulse width and PSO-BP-PID control.The results show that the subjects have more stable riding speed and longer continuous training time in the FES rehabilitation bike control system.