Research On Trajectory Planning And Compliance Control Of Autonomous Assembly Robots

With the diversification and complexity of market demand,the complexity of the environment,the accuracy of workpiece positioning,and the diversification of assembly objects in assembly operations have caused great challenges to the autonomous and precise assembly of robotic arms.Therefore,it is of great significance to study the autonomy and intelligence of robot assembly operations.This article has carried out related research on the algorithm of axis hole autonomous assembly,the main contents are as follows:(1)Efficient trajectory planning of the robotic arm.Faced with the manifold characteristics of the orientation and the zero-crossing situation,the existing dynamic movement primitives(DMPs)algorithms are difficult to achieve the expected results.This paper proposes a 6D trajectory planning method in Cartesian space based on improved DMPs.This method uses quaternion to describe the orientation,and realizes the non-singular representation of the position trajectory and the orientation trajectory.By decoupling the correlation between the forcing function and the start-end state difference term,the problems of trajectory jitter,inability to generate and flip caused by the zero crossing point are eliminated.In addition,a virtual impedance relationship is established based on the distance and deflection angle between the manipulator and the obstacle,and it is coupled to the dynamic model,making the obstacle avoidance behavior of the robotic arm more human-like and reducing consumption.Simulation and pouring experiments show that the improved DMPs method proposed in this paper is feasible.(2)Compliance control based on impedance.In the contact task,the pure position control of the robot arm in the past has been difficult to meet the requirements,and the simultaneous control of the position and force between the robot arm and the environment can be achieved through impedance control.However,when faced with an unstructured operating environment,the impedance control with fixed parameters is often difficult to achieve the desired effect.For this reason,this article enriches the impedance model by means of quaternion,which is conducive to the dual-compliant control of position and posture.Secondly,the iterative learning control method is used to optimize the output of the impedance controller,which improves the response performance of the system.Finally,a simulation system was built through Matlab/Simulink,and the experiment verified the effectiveness and robustness of the algorithm.(3)Robotic arm assembly operation.Based on the ROS framework,a real-time control system is developed to facilitate real-time feedback control of the assembly process.Analyze the realization process of tool coordinate system calibration,force/torque sensor calibration and gravity compensation in the assembly process,so as to facilitate the conversion and acquisition of various states in the assembly operation.Finally,the free drag and peg insertion experiment of the UR5 manipulator is designed to verify the feasibility of the real-time control system and algorithm.