Research On Self-Calibration Method For Industrial Robots

Manufacturing industry is the cornerstone of national economy,supporting the stable development of the country in different periods.In the environment of promoting intelligent manufacturing,industrial enterprises have an increasing demand for.applications where the robot should be programmed through off-line programming.However,at present,industrial robots normally have high repeatability(about 0.1mm or better)but low positioning accuracy(several millimeters or worse),which has become an important factor limiting the applications of industrial robots,especially in offline programming.Kinematic calibration serves as an effective solution to improve the positioning accuracy of an industrial robot.The traditional robot calibration method uses expensive and cumbersome external measuring equipment to obtain the robot pose,which has the disadvantages of high cost and poor portability.Self-calibration methods utilize geometric constraints and sensors to obtain the robot pose,which can overcome most of the shortcomings of traditional calibration.Generally,an ideal self-calibration method should have a complete and parametrically continuous calibration model as well as a low-cost,large-workspace and reliable-accuracy calibration device.In order to solve the problems mentioned above,this paper presents a portable and low-cost self-calibration device,which is composed of a spherical coordinate measuring device installed at the end of the robot,a Tri-ball device fixed relative to the robot base and a movable double ball device.By using spherical constraint,point constraint and distance constraint respectively,the distance error of the robot can be measured in a large working space and the calibration can be improved At the same time,it can measure the position error of the robot in the local workspace,and define the external reference coordinate system which is used to describe all the coordinate systems in the robot working cell.Based on the designed device,double ball calibration algorithm based on distance error,Tri-ball calibration algorithm based on position error and two-step self-calibration algorithm combining distance and error information are proposed respectively.In the first step,the proposed two-step calibration method uses the distance error information in a wide range of workspace to calibrate the robot.Subsequently,in the second step,the user-defined coordinate system is established,and the robot base coordinate system is calibrated using the position error,and the calculation results obtained in the first step are further improved;finally,an accurate motion model relative to the user-defined coordinate system is established It is a relatively ideal calibration method to improve the accuracy and effectiveness of off-line programming applications.For the three proposed calibration algorithms,different from the previous scholars’ application of the local index formula in the calibration,this paper adopts the local index product formula of the screw displacement type,and introduces the position adjoint transformation matrix,and establishes a self-calibration model with further simplification of form,linear unification of structure,and the same time error parameters meeting the requirements of completeness and continuity.Finally,the self-calibration algorithm is simulated,and the simulation results show that the double ball calibration algorithm and the two-step self-calibration algorithm proposed in this paper can quickly converge to the magnitude of noise amplitude in the case of noise,which verifies the effectiveness and robustness of the calibration algorithm.Based on the designed device and algorithm,a robot self-calibration system is built,and its experimental research is carried out.The experimental results show that the self-calibration measurement device has a certain practicality,and also verify the effectiveness of the proposed two-step calibration algorithm and Tri-ball calibration algorithm in the actual calibration experiment of robot position error compensation.