A mapping method and device for a global visual coordinate system and a robot arm coordinate system

Abstract

The invention provides a global visual coordinate system and mechanical arm coordinate system mapping method and device. The method includes the following steps that: S1, the first ball center coordinate of the ball center of a target ball on the distal end of a mechanical arm under a mechanical arm coordinate system is acquired; S2, image recognition is performed on the target ball, the ball center is extracted from the image through a ball detection algorithm, and conversion is performed, so that the surface point coordinate of the target ball under a visual coordinate system is obtained; S3, the second ball center coordinate of the target ball in the visual coordinate system is calculated according to the surface point coordinate and radius of the target ball; S4, the position of the target ball is adjusted, the step S1 is executed again until a preset number of groups of first ball center coordinates and second ball center coordinates are obtained; and S5, and the preset number ofgroups of first ball center coordinates and second ball center coordinates are calculated through the least squares method, so that a mapping relationship between the two coordinate systems can be obtained. With the method and device of the invention adopted, automatic calculation can be realized efficiently and quickly, and mapping conversion precision is high.

Description

Technical field [0001] The present invention relates to an information processing robot, and in particular relates to a method for mapping the global coordinate system and a visual coordinate system and the robot arm means. Background technique [0002] With social progress and development of science and technology, the robot in the production of human life more and more widely. Robot itself has a separate set of world space, but in practical application scenarios, the need to use other equipment to get the world coordinate system to establish a working space, independent of each other between the two sets of coordinates, when you want to complete robotic automation Some specific needs, you will need two coordinate systems were unified. [0003] Currently, the robot coordinate system conversion mapping algorithm, it requires manually by sampling the common point of the two four spatial coordinate system modeling and measurements, human error inevitably exists, is very inefficient...

AI Technical Summary

Problems solved by technology

[0003] At present, the robot coordinate system mapping transformation algorithm can sample four common points in the two coordinate systems, but it requires manual modeling and measurement, and there are inevitably human errors, which is very inefficient

Images