Research On Force/Position Compliance Control Of Industrial Robot Based On Dynamic Parameter Identification

With the increasing difficulty and accuracy requirements of robot machining,multi-objective optimization control is becoming more and more important.In particular,when it comes to tasks requiring interaction with the environment,such as assembling and polishing,there might be excessive joint torque or surface damage of the workpiece,making it necessary to carry out compliant control on force and position.At the same time,given that the link structure and the motion state of the robot affect the force,a dynamics model must be established to analyze the relationship between the motion state and the force of the robot,in order to achieve the high-precision compliance control of force/bit.Aiming at the problems above,a method based on dynamics models to realize the end force/position compliance control of robots is proposed.The main research contents include:The commonly used robot dynamics modeling method is analyzed and compared.The Newton-Eulerian method is used to establish the robot dynamics expression considering friction model.Then the robot dynamics parameters are divided into the link dynamics parameters and the load dynamics parameters.In the dynamics parameter identification of the connecting rod,linear friction force is calculated.Based on the no-load dynamics model,the dynamics parameter set of connecting rod to be identified is established.And an improved genetic algorithm is proposed to identify the dynamics parameter set of connecting rod,which can effectively increase the convergence speed of identification.In the dynamics parameter identification of load dynamics,the expression method of the theoretical load torque of the robot with load dynamics parameters is derived.The load dynamics parameters are identified by the least squares method and the improved particle swarm optimization algorithm respectively.In contrast,the improved particle swarm optimization algorithm has higher recognition accuracy.Based on the dynamic model,the force/bit compliance control algorithm is studied.The position-based and torque-based impedance controllers are designed respectively.And The torque-based impedance controller is used to realize the coordinated control of the end force and position of the robot.The force/position compliance control of the designed impedance controller under external torque is verified by simulation.Finally,the dynamics parameters of the connecting rod and the load are identified via six-link robot experimental platform,and the impedance control algorithm is verified.A position error of up to 0.05 m is generated when the robot passes the obstacle trajectory,which confirms the practicability and reliability of the impedance controller based on the dynamic model.