| In recent years,increasing number of patients with limb motor dysfunction due to diseases such as spinal cord injury and stroke,and exercise rehabilitation therapy is very important for the recovery of motor function.Rehabilitation robots can coordinate patients for rehabilitation exercises with high precision,but for most people who lose their ability,they can only achieve passive auxiliary actions,and have limited muscle activity.Functional electrical stimulation technology mainly uses low-frequency weak current pulses to stimulate the muscles of paralyzed patients,allowing muscles to contract to replace or correct the function that limbs have lost.However,because the human musculoskeletal system is a highly nonlinear time-varying system,combined with the influence of disturbance factors such as muscle fatigue,the control accuracy of the functional electrical stimulation control system is reduced after repeated movements,and the expected movement cannot be completed.Therefore,the combination of functional electrical stimulation and rehabilitation robots has become a new rehabilitation technology.Through functional electrical stimulation,the muscle activity is improved as much as possible,and the patient's own muscle contraction is used to complete the rehabilitation training which can improve the patient's enthusiasm for participating in training.In order to improve the control precision of the repetitive motion,then the human body lower limbs can accurately track the predetermined training trajectory.In this paper,the lower extremity knee joint and ankle joint are studied,and the muscle model of the lower limb joint under electrical stimulation is studied.A functional electrical stimulation control method for lower limb rehabilitation is proposed,and a flexible muscle-driven flexible ankle rehabilitation robot is integrated,so that the functional electrical stimulation and the ankle rehabilitation robot work together on the ankle joint to complete the rehabilitation training track.The main research contents of this paper are as follows:(1)According to the characteristics of human lower extremity joint movement and functional requirements,construct a joint kinematics model of lower extremity,analyze the muscle response characteristics under functional electrical stimulation,introduce Hammerstein-ARMAX model,establish the dynamics of functional electrical stimulation pulse width and electrical stimulation muscle torque.The relationship between the static nonlinear part and the dynamic linear part of the model is identified by particle swarm optimization algorithm.The correlation between pulse width and electrical stimulation torque is analyzed by experiments to verify the feasibility and effectiveness of the overall model.(2)Considering that multiple muscles need to be coordinated in the movement of multiple joints in walking,each joint also has coupling,and a decentralized control framework is proposed.An independent system is designed for each joint,and regards the interaction between subsystems as external interference.Each independent system adopts adaptive sliding mode control.At the same time,for the nonlinear and timevarying characteristics of the model under functional electrical stimulation,the RBF network is used for approximation,and a single parameter is used instead of the weight in the neural network.The adaptive control of the parameter estimation compensates for external disturbances,so that the functional electrical stimulation generates a boosting effect during the movement of the lower limbs of the human body,and achieves accurate tracking of the motion trajectory.(3)For people with impaired exercise ability,rehabilitation robots have limited ability to improve their muscle activity,while long-term stimulation may easily lead to muscle fatigue.Therefore,in order to improve the degree of muscle activation during the rehabilitation process and avoid muscle fatigue,the two are combined to assist the patient to complete the ankle joint rehabilitation training,and at the same time,the functional electrical stimulation is used to stimulate the knee joint,and the multi-joint exercise simulating the gait training is completed together.In order to realize the flexibility of human-computer interaction,the impedance model of the ankle rehabilitation robot is established.At the same time,considering the repeated training of the patient,it is easy to cause muscle fatigue.According to the principle of energy minimization and muscle fatigue minimization,a dynamic allocation control method is proposed to optimize the input of functional electrical stimulation and coordinate the distribution of electrical stimulation and torque generated by the robot.In this way,the completion of the overall movement can be ensured,and the activity of the muscle can be fully utilized to achieve optimality,so that the patient can perform the desired exercise under the assistance of the functional electrical stimulation and the robot. |
PDF Full Text Request