Research On Calibration Of Kinematic Parameters Of Manipulators

Robot positioning accuracy is an important index to evaluate the comprehensive performance of the robot,which is used to measure the accuracy of the robot end effector reaching the expected target in the work space.The positioning accuracy of the robot can be divided into the repeat positioning accuracy of the end effector and the absolute positioning accuracy of the end effector.In current production applications,robots have high repetitive positioning accuracy,but low absolute positioning accuracy.The kinematic parameter calibration method based on the four steps of modeling,measurement,identification,and compensation has a significant effect in improving absolute positioning accuracy,and has low requirements for mechanical structure hardware,strong adaptability,and easy application.Taking the six degrees of manipulator as the research carrier,the absolute positioning accuracy of the robot is improved by studying the calibration method of kinematic parameters,and the application value of the robot is improved.The specific research content is as follows:(1)Carry out kinematics analysis on the manipulator,use the Denavit-Hartenberg modeling method to establish the kinematics model of the six degrees of manipulator,derive the forward kinematics equation of the manipulator,and complete the verification of the model in the Matlab robot toolbox.On this basis,the Modified-Denavit-Hartenberg modeling method is used to avoid the problem of model parameter mutation when the adjacent joint axes of the manipulator are parallel or close to parallel,and the end position error model of the manipulator is established in combination with the differential motion transformation theory.(2)Propose and implement two identification methods to solve and compensate the error model.Firstly,the solution and compensation of the error parameters of the end position error model of the manipulator are realized by the least square method.Considering that there will be errors in directly solving the identification model,and the obtained approximate solution,a hybrid heuristic algorithm is proposed to identify and compensate the error model,that is,an adaptive genetic simulated annealing algorithm based on the combination of genetic algorithm and simulated annealing algorithm.The algorithm uses adaptive genetic algorithm as the main framework,and uses simulated annealing algorithm to improve the local search ability of adaptive genetic algorithm.Firstly,the traditional genetic algorithm is improved,and an adaptive thinking is used to solve the problem of calibration.The adaptive algorithm improves the overall robustness of the algorithm by adaptively adjusting the crossover and mutation probabilities of the algorithm,so that the algorithm shows better performance in solving the problems,introduce the simulated annealing algorithm to guide the local refinement of the adaptive genetic algorithm.Finally,through the design simulation scheme,it is verified in the Matlab simulation environment that the adaptive genetic simulated annealing algorithm balances the ability of algorithm exploration and development,and improves the solution accuracy of the error model of the manipulator.(3)Finally,the kinematic parameter calibration experiment was successfully carried out based on the six degrees of manipulator ABBS6 robot.Firstly,the experiment scheme is designed,and the kinematics model and the terminal parameter error model are established according to the structural analysis of the ABBS6 robot;the common working space is calculated by the Monte Carlo method,and the terminal position error is measured by the Leitz laser range finder;the minimum The quadratic method and the adaptive genetic simulated annealing algorithm are used to identify and compensate the error parameter model.Finally,the calibration results of the two identification methods were analyzed and compared,and the experiment was successfully completed.The average distance error between the theoretical position of the end of the experimental sample data and the actual position is reduced from 4.79 mm before calibration compensation to 1.33 mm and 0.53 mm respectively,and the absolute positioning accuracy has increased by 72.3% and 89% respectively.The results of actual measurement show that after kinematic parameter identification and compensation correction,the absolute positioning accuracy of the end effector of the manipulator is significantly improved,which verifies the effectiveness of the proposed kinematic parameter calibration method.And compared with the commonly used least square method for identification and compensation,the proposed adaptive genetic simulated annealing algorithm has better identification and compensation results.