Transformation error compensation method for robot flexible joints based on improved pi structure

Abstract

The invention discloses a robot flexible joint conversion error compensation method based on an improved PI structure. On the basis of the structure of the symmetrical Play operator, the linear part in the Play operator is replaced by an improved nonlinear Sigmoid function, and a hysteresis is constructed. A new function close to the curve profile is obtained to obtain a nonlinear hysteresis operator. Using the new hysteresis operator as an activation function, a neural network hysteresis model is constructed to model the complex hysteresis characteristics of flexible joints, and based on the neural network hysteresis model The control compensation of the drive motor of the flexible joint is performed. The hysteresis model of the neural network of the present invention has the ability of online learning, and can compensate the non-linear error of transmission caused by the structure of the joint itself of the industrial robot on line, so as to improve the execution accuracy of the joint of the industrial robot.

Description

technical field [0001] The invention relates to the technical field of industrial robots, in particular to a method for compensating conversion errors of robot flexible joints based on an improved PI structure. Background technique [0002] Industrial robots are more and more widely used in the field of industrial production. In the process of intelligent manufacturing, the requirements for precise control of industrial robots are getting higher and higher. Carrying out modeling and compensation of strong nonlinear characteristics of robot joints has become an important technical way to improve the control accuracy of industrial robots. [0003] In order to improve the interaction and cooperation ability between the robot and the environment or human beings, the use of flexible materials or flexible transmission elements can reduce the impact during the interaction process and ensure safety. For collaborative robots and light robots, the flexible joints that contain harmoni...

AI Technical Summary

Problems solved by technology

However, the particularity of the complex nonlinearity exhibited by the flexible joints of industrial robots lies in its asymmetry and non-smoothness. The traditional PI model is suitable for the description of the symmetric hysteresis curve. For complex hysteresis characteristics, the modeling accuracy is reduced by using the traditional PI structure.

In recent years, the method of modeling hysteresis characteristics of objects has been improved by improving the PI structure, such as the use of variable interval thresholds and the use of three-segment PI modeling, but these improvements only broaden the traditional PI model and cannot fundamentally solve the problem of industrial robots. Hysteresis properties exhibited by flexible joints that are asymmetric, non-smooth, and without absolute concavities and convexities

Images