Industrial robot positioning precision calibration method, device and system

Abstract

The invention discloses an industrial robot positioning precision calibration method, device and system, and the method comprises the steps: S101, respectively obtaining the actual positions and theoretical positions of a robot at n sampling points; S102, performing difference processing on the actual position and the theoretical position to obtain error values of the robot at n sampling points; S103, taking the theoretical position of the robot as an input layer, taking an error value corresponding to the theoretical position as an output layer, training a hidden layer by adopting an optimalpruning extreme learning machine model, and establishing an error model; S104, inputting a target position to be reached by the robot into the error model, and calculating a deviation value of the target position; S105, superposing the target position to be reached by the robot with the compensation value corresponding to the position to obtain an expected target position of the robot. Motion parameters of the industrial robot control system do not need to be modified, the development degree of the industrial robot control system is reduced, and the use cost of enterprises is reduced.

Description

technical field [0001] The invention belongs to the technical field of calibration of industrial robots, and in particular relates to a positioning accuracy calibration method, device and system of an industrial robot. Background technique [0002] With the gradual deepening of the development of intelligent manufacturing technology, industrial robots are used more and more in product manufacturing, processing and assembly, deburring and polishing, offline programming, etc. These applications mainly rely on the absolute positioning accuracy of robots, and industrial robots generally have relatively high High repeatability positioning accuracy, while the absolute positioning accuracy is low, and the low absolute positioning accuracy is difficult to guarantee product quality, so it is particularly important to improve the absolute positioning accuracy by calibrating the robot. [0003] At present, the commonly used methods for robot calibration are to establish robot error mod...

AI Technical Summary

Problems solved by technology

[0002] With the gradual deepening of the development of intelligent manufacturing technology, industrial robots are used more and more in product manufacturing, processing and assembly, deburring and polishing, offline programming, etc. These applications mainly rely on the absolute positioning accuracy of robots, and industrial robots generally have relatively high High repetitive positioning accuracy, but low absolute positioning accuracy, it is difficult to guarantee product quality with low absolute positioning accuracy, so it is particularly important to improve absolute positioning accuracy by calibrating the robot

[0003] At present, the commonly used methods for robot calibration are to establish robot error models, pose measurement, parameter identification, and error compensation calibration methods, which not only require high technical level of users, but also require compensation of kinematic parameters in the robot control system after calibration. The control systems of most industrial robot manufacturers are not open to the public or require additional costs, so this method is difficult to promote on the production site

Images