Kinetic parameter identification method for SCARA

Abstract

The invention relates to a kinetic parameter identification method for an SCARA, which corrects a model of a robot through kinetic parameter identification in order to obtain an accurate kinetic model for realizing accurate control of the robot. The invention belongs to the fields of system identification and robot control. The method comprises the following steps of S1, conducting D-H modeling on the robot, obtaining parameters of each axis, establishing a robot kinetic model, and conducting linearization; S2, determining a minimum inertial parameter set, and obtaining a matrix equation formed by an observation matrix, to-be-identified parameters and the torque; S3, designing an excitation track with good properties to act on the robot, measuring related data and conducting noise reduction processing; and S4, substituting the actual measurement data into a differential evolution algorithm to conduct kinetic parameter identification. According to the method, the convergence speed is high, the calculation cost is relatively low, all joint friction and other kinetic parameters can be identified at a time, and the method has relatively high identification precision.

Description

technical field [0001] The invention relates to the technical field of robot control and system identification, in particular to a dynamic parameter identification method for a SCARA mechanical arm. Background technique [0002] As we all know, automated robot systems have been widely used in various industrial places, and the various application scenarios of robots also put forward higher requirements for control accuracy. Based on this demand, we need to obtain more accurate robot power Parameter identification is the most common method to solve this problem. [0003] According to investigations, the currently commonly used methods for identifying robot dynamic parameters include physical experiment method, CAD measurement method and least square method, etc. [0004] The physical experiment method mentioned above is to measure the parameters of the robot directly through the experiment. Obviously, this method cannot consider the coupling and friction inside the robot, an...

AI Technical Summary

Problems solved by technology

[0004] The physical experiment method mentioned above is to measure the parameters of the robot directly through the experiment. Obviously, this method cannot consider the coupling and friction inside the robot, and has great defects; while the CAD drawing can only show the robot parameters. In the actual operation of the robot, there must be a large error between it and the parameters of the actual model; the disadvantage of the least squares method is that the data that needs to be collected and processed is huge, the identification cost is high, and it cannot be dynamically adjusted according to the increase of identification data. renew

Images