Research On Trajectory Planning And Compliant Control Of Lower Limb Rehabilitation Robot

Lower limb rehabilitation robot can provide physical rehabilitation training for patients who suffer from stroke or hemiplegia caused by accidents. It is effective to improve the dexterity by rehabilitation robot, for it can realize precise motion control, training and online assessment. Parallel robot of six DOFS, due to advantages in large stiffness, small cumulative error, high positioning accuracy and strong load capacity, have been widely used in the space docking movement, vibration isolation, driving simulation, medical equipment, etc. In the field of lower limb rehabilitation, on the one hand parallel robot can liberates rehabilitation physicians from the heavy manual training, and on the other it provides scientific training for patients to restore motion function.This paper studies the trajectory planning, high precision position control and compliant force control in the process of lower limb rehabilitation, which can realize effective rehabilitation training and obtain multiple rehabilitation modes. The research content mainly contains the following parts:(1)In order to obtain the human gait data of knee, ankle and foot, the mathematical model of gait information is established based on the video capture system. Cubic spline function is used in the trajectory planning, which ensures the smooth transition in trajectory speed and acceleration. So it can prevent secondary damage to patients in the rehabilitation training. The information of trajectory speed and acceleration in the joint space is obtained by the inverse kinematic of parallel robot.(2)The position control of the parallel rehabilitation robot is researched. In order to ensure the accuracy of gait trajectory tracking in the rehabilitation training and improve the stability of the system, the nonsingular terminal sliding mode control is adopted. Combined with the specific parameters of the parallel robot and the idea of the exponential reaching law, the paper puts forward suitable control law, under which the stability of the parallel robot system is proved through the Lyapunov stability theorem and the trajectory tracking error is guaranteed to convergence to zero in the finite time.(3)The compliant control of the parallel rehabilitation robot is researched. The admittance control is proposed to do compliant rehabilitation training. The paper analysis the three key parameters in the proposed admittance control and discusses their effects on the performance of force control, based on which the variable damping admittance control is proposed. Using sigmoid function to change the damping parameter, it can improve the transient performance and prevent overshoot in the force control. The proposed method ensures the safety and prevents the mutation and overshoot of interaction force in the rehabilitation.According to the above research, the human normal gait information and ideal rehabilitation training trajectory can be obtained effectively. Simulation and experimental results show that the nonsingular terminal sliding mode control can effectively ensure the accuracy of trajectory tracking and the system stability, and variable damping admittance control has the response rapidity and small overshoot in force control, and moving platform can adjust the actual trajectory effectively according to the man-machine contact force.