Industrial Robot Calibration And Experimental Research

With the continuous expansion of robot application,in order to meet the high-precision requirements in industry production,to improve the robot's absolute positioning accuracy has become one of the key technical issues.In this paper,taking UR3 as a research subject,adopting Kinematic calibration methods based on D-H parameters,using laser tracker measuring system to identify the accurate parameters and increase the absolute accuracy.Based on the calibration of nominal kinematic parameters,the function mapping between the position error of end effector and the load was also studied.The prediction function of the position error caused by the load was established and the position accuracy was compensated by driving the robot joints small angles to further improve the position accuracy.The main research contents are as follows:(1)Combining with the features of UR3 robot,based on the D-H parameter method and the POE formula,the kinematic models of UR3 industrial robot were established respectively.The advantages and disadvantages of the two methods were analyzed and compared.The human-machine interface was written with MATLAB language,it was verified that the forward algorithm written in MATLAB language were consistent with the algorithm in UR3 controller.(2)In this paper,we derived the linear error equation of the robot by using the differential method.By using QR decomposition of the identification Jacobian matrix,the optimal measuring pose set was obtained by adopting the first type of observability index from the candidate measuring pose set.UR3 robot calibration program and the optimal measurement pose program were written in MATLAB language.(3)Based on the calibration method of UR3,calibration experiment platform was set up,and the pose of the robot end-effector were measured by a laser tracker.The LM algorithm was used to identify the kinematic parameters.Finally,the result of the parameter deviation was compensated to the robot controller.The comparison of position error before and after calibration showed that the average absolute position accuracy of robot has been significantly improved(4)According to the measurement data,the relationship between the position error of the end effector and the joints rotation angles under static load was established based on Fourier polynomial and ordinary polynomial.The position error prediction functions of single joint motion were weighted,and the position error prediction function of robot end effector to static load in multi-joint motion were obtained.The position error predicted by the position error prediction function of the robot end effector was used to compensate the position of the end effector.The experimental results verify the effectiveness of the method.(5)The position error of the end effector of UR3 robot under different static loads was studied.Combined with the features of UR3 robot structure,the influence of different static loads on robot position accuracy was analyzed.The position error prediction model of UR3 robot end effector caused by different static loads was proposed.