Robot non-redundancy geometric error model analysis modeling method based on spinor theory

Abstract

The invention discloses a robot non-redundancy geometric error model analysis modeling method based on a spinor theory. The method comprises the steps that 1, a series branch chain full-parameter geometric error model is established; 2, a parallel kinematic chain full-parameter geometric error model is established; and 3, redundant geometric error parameters in a parallel kinematic chain are deleted, and the number of maximum recognizable geometric error parameters of the parallel kinematic chain is solved. The method is not only suitable for geometric error modeling of parallel configuration robots, but also suitable for geometric error modeling of series-parallel hybrid configuration robots, and is suitable for geometric error modeling of robots of any configuration; and a model established by the method has no redundant error parameters, can explicitly acquire all mutually independent geometric error parameters with clear physical significance in kinematic chains of various types of robots, and further can guide precision design and kinematic calibration work of the robots at the same time.

Description

technical field [0001] The invention belongs to the field of robots, in particular to an error modeling method for a parallel configuration robot and other mechanical equipment. Background technique [0002] At present, there are two main technical means to improve the robot’s accuracy: one is to ensure the basic manufacturing accuracy of the robot through precision design, and the other is to further improve the robot’s end pose accuracy through kinematic calibration. Whether it is precision design or kinematics calibration, it is necessary to establish a robot model without redundant geometric errors. Geometric error is an important factor affecting the end accuracy of robots and other mechanical equipment. By suppressing the influence of geometric error sources on the end pose error, the accuracy performance of the system can be effectively improved. Geometric error modeling aims to establish the linear mapping relationship between the robot end pose error and the geomet...

AI Technical Summary

Problems solved by technology

Existing modeling methods for robot geometric errors include vector method, D-H method, homogeneous coordinate transformation method, screw method, exponential product method, CPC method, etc., but there are two types of limitations: the existing modeling methods cannot effectively deal with mechanism Due to the orthogonal duality relationship between the subspaces of the motion space and the force space, it is difficult to obtain a geometric error-free model by analytical means, so the model cannot be directly used for robot precision design and kinematics calibration; When reducing the redundant error parameters in the model, the error parameters of the robot drive joint motion may be deleted by mistake, and the deleted model often loses its clear physical meaning, so it cannot be used for the precision design and kinematics calibration of the robot

Images