Research On Path Planning And Force Control Of Peg-Hole Assembly Robot

As an indispensable link in the manufacturing process of a product,assembly is also the most labor-intensive process.However,existing assembly robots have the disadvantages of low intelligence,poor adaptability,and low efficiency,which increase the intelligence of assembly robots,It is of great significance for China to realize"Made in China 2025."Based on the traditional industrial manipulator,this paper studies the path planning,force control,and assembly system,and lays the foundation for realizing the robot's autonomous peg-hole assembly.1.path planning.From the grabbing of the workpiece to the pre-installed position,a robot's autonomous planning path is required.First,under the premise of taking into account the computing power of the computer,the rapidly-exploring random tree(RRT)algorithm is selected as the basis,and the disadvantages of greedy growth and slow convergence are improved.By adding the target gravity function,the average time of the algorithm is reduced from the original 1.16s to about 0.3s.Secondly,taking into account the link constraint of the robot,the obstacles are uniformly mapped to the joint space of the robot,the improved RRT algorithm is used for path planning in the joint space,and the planned path is pruned and smoothed so that It is meet with the robot's kinematics requirements.Finally,the robotic toolbox is used to build the simulation model of the robot under the matlab,and the path of the simulation proves the feasibility and effectiveness of the method on the robot path planning.Experiments were carried out.The path generated by the RRT algorithm was sent to the motion controller through the measurement and control system.The experiment verified that the path can effectively avoid obstacles in the space.2.Sensor-based peg-hole relative pose measurement.First,the position and orientation deviation of the peg-hole at the three-point-contact are described.Secondly,through the force analysis and geometric analysis of the peg-hole under the three-point-contact,the relationship between the peg-hole position deviation,orientation deviation and force is deduced.Finally,the data of the six-axis force sensor can be used to roughly deduct the deviation of the position and orientation,which in turn helps the robot to leave the three-point-contact state.3.Robotic impedance force control based on fuzzy rules.First,consider that the data collected by the sensor often has Gaussian white noise,and the kalman filter can effectively filter the noise,which can improve the stability of the force control system.Second,establish an impedance model between the robot and the external environment,and use the impedance characteristics to control the force of the robot.through the fuzzy rules,the force control system automatically adjusts the parameters of the impedance.The force error and the variation of the error are taken as the input of the fuzzy controller and the adjustment of the impedance parameter is taken as the output.Finally,Simulink is used to construct the force control system simulation platform,and the robot's hole searching process is simulated under the platform,so that it can track the circular trajectory on the XY plane and keep constant force or variable force tracking in the Z direction.Experimental results show that in the case of impedance control with fuzzy rules added,the steady-state force tracking error is reduced from±1N to±0.1N in both the case of constant force and variable force,and the response speed is faster and overshoot is smaller.4.Assembly robot control system development.First,according to the requirements of the assembly robot,an overall design of its control system is made.The control system is decomposed into mutually independent sub-modules.Then made a brief introduction to the entire hardware and software platform of the robot system.a detailed description of the communication module and the secondary development of the force sensor is made.Finally,based on the Visual Studio development environment,the assembly robot control system was completed.