Research On Trajectory Control Of Human Knee Joint Based On Functional Electrical Stimulation

In the current society,stroke occurs frequently.Functional electrical stimulation(FES)has become a major measure of functional rehabilitation for stroke patients because of its unique rehabilitation effect.The function of joint trajectory control based on functional electrical stimulation can not only promote functional rehabilitation of patients,but also help to enhance patients' confidence in rehabilitation.Functional electrical stimulation takes human muscles as actuators,and human muscles have nonlinearity and time-varying characteristics,so it is difficult to have accurate model expression.As a result,the dynamic feedforward accuracy of human joint trajectory control is not high enough.The problem of inadequate dynamic feedforward accuracy can only be remedied by artificial intelligence algorithms such as fuzzy control and closed-loop control.Therefore,the human body based on functional electrical stimulation will be developed.Joint trajectory control combines with human musculoskeletal modeling,using precise human musculoskeletal model to improve the accuracy of dynamic feedforward,thereby improving the control accuracy,which becomes an achievable direction.At the same time,most of the current trajectory control of human joints based on functional electrical stimulation is single muscle control,which can not achieve multi-muscle coordinated control,and there will be unreasonable stimulation mode.Therefore,this topic uses multi-muscle coordinated stimulation.For most human joints,the degree of freedom of joints is generally less than the number of muscles that cause the joint movement,which is mechanical redundancy.Therefore,there is a problem of muscle force optimization in multimuscle coordinated stimulus on the basis of satisfying kinematic trajectory constraints.Aiming at the problems of low accuracy of dynamic feedforward control of human joint trajectory control based on functional electrical stimulation and mechanical redundancy of multi-muscle coordinated stimulation,this paper focuses on the research of human lower limb musculoskeletal modeling and electrical stimulation simulation and muscle force optimization,and integrates them into the control of knee joint trajectory tracking.The joint trajectory is verified by simulation and experiment.The correctness and effectiveness of the control method.Specifically,the following studies are carried out:Firstly,in order to establish an accurate lower limb musculoskeletal model,the motion capture system is used to collect human marker information.By comparing with the general musculoskeletal model,the specific musculoskeletal model for the experimenter is obtained.Through the study of muscle activation dynamics and muscle contraction dynamics,the mapping relationship between electric pulse parameters and muscle force is established,and the MATL is built.The AB/opensim joint simulation platform simulates the characteristics of muscle force generation of a single muscle,and obtains the physical model of muscle force with the input of electric pulse parameters and the output of the output.Secondly,aiming at the problem of mechanical redundancy in multi-muscle coordinated stimulation,the joint moment is calculated by inverse kinematics and inverse dynamics.Several popular methods of muscle force optimization are compared and analyzed.CMC algorithm is selected to optimize muscle force.The principle and characteristics of CMC algorithm are analyzed.The muscle force optimization simulation is carried out by using CMC algorithm to obtain the muscle force.Muscle activation.On this basis,according to the characteristics of nonlinearity and time-varying of musculoskeletal system,the closed-loop control method of knee joint trajectory based on dynamic feed-forward model is proposed by using the precise musculoskeletal system and muscle force optimization method established before.The simulation results show that the control accuracy of the closed-loop control method is higher than that of open-loop control.Finally,in order to verify the validity and practicability of the control method,the parameters of functional electrical stimulation pulse are selected,the technical indicators of functional electrical stimulation experimental platform are determined,and the functional electrical stimulation experimental platform is built.Combined with the data acquisition system,the experimental platform of knee joint trajectory tracking control is established.Firstly,the trajectory tracking control experiment of single-subject knee joint is carried out to verify the effectiveness of the control method.Secondly,the multi-subject difference experiment is carried out to prove the applicability of the control method to different individuals.Finally,the influence of the range of motion on the control accuracy is tested by experiments.