Research On Impedance Control And Motion Planning Of Manipulator System With Uncertain Parameters

Impedance control is one of the key technologies for force/position tracking control of manipulators.However,traditional impedance control relies too much on accurate environmental model,and it is difficult to adapt to the working environment where the parameters cannot be accurately measured;in addition,modeling errors are unavoidable,which makes it difficult for the control methods based on known parameters to achieve high-precision trajectory tracking,which in turn affects the performance of the force/position tracking of the end of manipulator.How to reduce the influence of the uncertainty of the manipulator and the unknown environment on the force/position tracking of the manipulator has always been a research hotspot in the field of control.This thesis takes impedance control as the research basis,and considers the uncertainty of the dynamic parameters of the manipulator as the main research content.At the same time,the obstacle avoidance motion planning of the manipulator in complex scenes is studied.(1)In view of the influence of obstacles in the environment on the safety of the manipulator,the obstacle avoidance path planning is based on the artificial potential field method.The defect of the non-optimal point of the target point is solved by improving the repulsive potential field function of the artificial potential field;the variable step size mechanism is introduced to avoid re-planning a new path by eliminating the collision way-points,which improves the efficiency of the artificial potential field method;The trajectory planning of discrete way-points improves the smoothness of the way-points.(2)Considering the problem that the planned reference trajectory in the impedance function is not ideal due to insufficient prior knowledge of the environment,which affects the force/position tracking performance of the end of the manipulator,a force tracking scheme based on adaptive impedance control is designed.The reference trajectory is updated online in real time according to the force tracking error,so as to solve the influence of the environmental modeling error and indirectly improve the force tracking performance of the manipulator.Facing the unstructured environment,the control scheme has more flexibility and applicability,and improves the versatility of the control system.(3)Aim at solving the problem that the force/position tracking performance is degraded due to the dynamic uncertainty of the manipulator system,and considering the inevitable factor of external disturbance and the friction of the joint actuator,a position controller based on an adaptive neural network is designed.The single-parameter neural network fits Cartesian space dynamics,solves the problem of kinematic parameter uncertainty,and estimates the square of the maximum singular value of the ideal weight matrix of the neural network online,so that only a single learning parameter needs to be updated,thereby reducing calculate the pressure and improve the real-time performance of the control system.Finally,combined with the analysis of Lyapunov stability theorem,the stability of the system is strictly proved.(4)In order to improve the engineering application value of the force/position tracking control system,the algorithm was applied to the control system of the high-degree-of-freedom Sawyer manipulator,which highly verified the effectiveness and practicability of the control system,and greatly speeds up the practical application of the intelligent controller,which is of great significance to the engineering application of the intelligent controller.