Research And Design Of Manipulator Compliance Control Based On Impedance Control

Under-actuated manipulator is a commonly used end-effector in contact with the operating environment of the robot.It can replace manual operations,such as grinding,picking and assembly,when completed with the manipulator arm.It is a research hotspot of intelligent control.Nowadays,the requirement of manipulator is not simple position control,but adaptive force/position tracking control,to complete more accurate grasping action.At the present stage,the structural design,transmission mode and drive modularization of the manipulator are becoming more and more mature.Researchers will pay more attention to the control system,which determines the adaptive degree and control precision of the under-actuated manipulator.The thesis focuses on the environmental parameter identification and the under-actuated manipulator compliant control system.Firstly,an under-actuated manipulator which can be used for multi-objective grasping is taken as the research object to carry out the corresponding theoretical analysis and verification of the compliant control system.The influence of constrained environmental parameters on the impedance control system is taken into account.In the thesis,the unknown environment is described as the second-order linear system with dynamic,and the environmental parameters are identified based on the RLS method using the measured contact force by the force sensor array and the derived displacement of the contact point,which improved the under-actuated manipulator perception of the constrained environment.Secondly,the control strategy of the system consists of two parts: one is the outer loop of impedance control and the other is the inner loop of position control.The equivalent impedance model between the end of the under-actuated manipulator and the constrained environment is depicted by the outer loop of impedance control,aiming at establishing the dynamic relationship between the contact force and the position.The contact force and displacement at the end of the under-actuated manipulator is affected by the different constraint-environments,which makes it necessary to match the appropriate target impedance parameters.Therefore,selecting the appropriate impedance parameters is the focus of the study to solve the problem of outer loop impedance control.In the thesis,an adaptive fuzzy impedance parameter tuning method based on inversion method is proposed,which can effectively complete the adaptive adjustment of impedance parameters.The nonlinear influence of the manipulator itself on the system is considered,especially the dynamics model parameters of the manipulator.Therefore,a position controller based on RBF neural network approximation is designed in the position inner loop to approximate the parameters of the dynamics model and improve the robustness of the system.Finally,the effectiveness of the impedance parameters adaptive adjustment and RBF neural network control system is verified by a simplified 2-DOF model.Based on the existing experimental platform of under-actuated manipulator,the software and hardware design of the control system are completed.The reliability of the control system is proved by the force/bit tracking control experiment of the under-actuated manipulator in the free/constrained space.