Self-adaptive modeling method for robot drive and control integrated system

Abstract

The invention relates to the technical field of robot drive and control integrated systems, and discloses a self-adaptive modeling method for a robot drive control integrated system. The method comprises the following steps of 1) obtaining a kinematic model and kinetic parameters of each joint of a robot according to a design model of the robot; 2) calculating an issued torque by means of a Newton-Euler method according to an inverse kinetic model and a current planned position, velocity and acceleration; 3) by means of a dimensionality reduction method, conducting dimensionality reduction onthe robot drive control integrated system into an N-order system to obtain a state equation of the system; 4) setting an adjustable self-adaption law, conducting optimization through a deep learning algorithm, and selecting an exponential acceleration and deceleration curve as an excitation trajectory. The kinetic parameters of the joints of the robot are considered, the dynamic performance of therobot is combined, integrated modelling is conducted on the robot drive and control integrated system, accurate matching of the model is achieved, and it is proved that the self-adaptive modeling method for the robot drive control integrated system is effective.

Description

technical field [0001] The patent of the present invention relates to the technical field of the integrated robot driving and control system, and specifically relates to the adaptive modeling method of the integrated robot driving and controlling system. Background technique [0002] At present, the servo control of the robot mainly depends on the adjustment of the output by the PID control. In view of the limitations of its parameter setting, by considering the robot kinematic chain as a whole as a control object, using a multi-stage system with excellent dynamic performance as a reference model, through the stable Based on the theory of self-adaptive matching, the deep learning algorithm gradually iteratively approaches, how to obtain the precise modeling model of the integrated robot drive and control system. [0003] In the existing technology, due to the limitations of experience, experimentation and parameter setting of PID control adjustment, it is difficult for the e...

AI Technical Summary

Problems solved by technology

[0003] In the existing technology, due to the limitations of experience, experimentation and parameter setting of PID control adjustment, it is difficult for the established model to gradually approach the performance of the reference model; the traditional PID controller model does not consider the electromechanical dynamics model of the robot The controller cannot be accurately combined with the dynamic performance of the robot, and the parameter identification and modeling calculation of the traditional robot dynamic model are very complicated, which is difficult to achieve in the engineering field

Images