Research On Bionic Leg Design And Control Strategy Based On Pneumatic Muscle

Due to the increasing number of amputees caused by war,traffic accidents,natural disasters and diseases,the demand for medical auxiliary equipment such as bionic leg prosthesis is also increasing.As an important research direction in the field of rehabilitation robots,bionic legs play a vital role in the recovery of patients ’ motor function.Traditional passive or semi-active bionic legs are usually only structurally bionic,which can only rely on the user ’s residual limbs to walk passively,and cannot provide active torque.There are problems of poor human-machine coordination,large physical consumption,and poor wearing experience after the user wears them.The active bionic leg provides active torque for each joint of the lower limbs through external driving components,so that the bionic leg can better complete the movement cooperation with the healthy lower limbs of the patient,and greatly improve the user ’s wearing experience.In this thesis,aiming at the problems of insufficient flexibility and poor structural bionics of the bionic leg system in the human-machine coupling system composed of active bionic leg and user,an active bionic leg prosthesis based on pneumatic artificial muscle is designed,and its motion mechanics characteristics and control strategy are studied.It mainly includes three aspects : structural design,driving mode,motion control simulation and experiment of active bionic leg.The specific research work is as follows :(1)Active bionic leg structure design.After analyzing the motion characteristics of human lower limb joints,the joint freedom and overall basic dimensions of the bionic leg are determined.From the perspective of human anatomy and bionics,the design criteria of the bionic leg are determined and the structural design of each part of the bionic leg is completed.After comparing the advantages and disadvantages of the three common driving methods,the pneumatic muscle is selected as the knee joint driving element and the size model and installation method are determined.Finally,the strength of the weak parts is checked under Solidworks software,and the strength is verified to meet the requirements.(2)Dynamic characteristics analysis of active bionic leg.Firstly,the dynamic model of bionic leg system is established based on Lagrange equation after the model is simplified,and the equal relationship between joint driving torque and motion angle,angular velocity and angular acceleration is clarified.Then,the dynamic analysis of the three-dimensional model of the bionic leg is carried out by Adams software.Taking the normal human joint motion curve as input,the bionic leg drive control simulation,gait walking simulation and drive torque simulation are carried out respectively.The rationality and shock absorption of the bionic leg structure are verified.According to the maximum driving torque of the knee joint,it is verified that the selected pneumatic muscle model meets the required torque requirements.(3)The establishment of single degree of freedom knee joint system model.Firstly,the structure of the pneumatic artificial muscle is analyzed,and the static mathematical model of the pneumatic artificial muscle is established by the analytical method.According to the relevant parameters of the selected pneumatic valve,the mass flow equation and the pressure dynamic equation of the pneumatic muscle working process are introduced into the joint dynamics model to establish the knee joint system model.Then,the system is reduced to a single input single output system by differential adjustment.After the system model is simplified,the system state equation for the subsequent controller design is obtained.(4)Research on control strategy of bionic leg knee joint.The stability control of the knee joint is the basis of the cooperative stability control of the bionic leg machine.Aiming at the nonlinear problem caused by the pneumatic muscle to the system,an inversion controller with good control effect on the nonlinear system is designed to control the joint.At the same time,in order to reduce the influence of the uncertain term neglected by the modeling and the unknown external disturbance,the neural network is used to approximate the unknown term,and the neural network inversion controller is designed.The trajectory tracking simulation experiment is carried out under Matlab.The results show that the RBF neural network inversion control can achieve stability control in the case of system nonlinearity and unknown interference.(5)Experimental study on bionic leg knee joint prototype.According to the designed bionic leg knee joint structure,the bionic leg knee joint parts were manufactured and the knee joint prototype experimental platform was built.In order to further verify the control effect of RBF neural network backstepping control on joints,joint trajectory tracking experiments under different tracking signals and different loads are designed to verify the effectiveness of the control method.