System and method for measuring the position and orientation of the end of the manipulator based on the color spherical light source target

Abstract

The invention discloses a robot arm end pose measuring system based on a colorful spherical light source target and a method thereof. The system comprises the colorful spherical light source target, alight source control module, a fixed mounted camera, a computer, a mechanical arm and a target calibrating system. The colorful spherical light source target comprises four vertically crossed supporting rods. The end of each supporting rod is provided with a set of light sources. Each set of light sources comprises at least one positioning center, and three light sources which are distributed along the positioning center. The colors of twelve light sources are different. The light sources are spherical. A supporting rod is mounted on the mechanical arm and moves along with movement of the mechanical arm. The target calibrating system is used for calculating the position of the light source target in a robot arm end coordinate system. The camera is fixedly mounted for acquiring an end motion image of the robot arm. The computer performs image subtraction, color correction, target identification and pose resolving for obtaining the pose of the robot arm end in the camera coordinate system.

Description

technical field [0001] The invention relates to a system and method for measuring the position and posture of the end of a mechanical arm based on a colored spherical light source target. Background technique [0002] Realizing the precise pose measurement of the end of the manipulator is of great significance for manipulator calibration and manipulator grasping. The current pose measurement method based on machine vision at the end of the robotic arm has the following problems: [0003] 1. Due to the large range of motion at the end of the robotic arm, there is an occlusion problem. When the end of the robotic arm is blocked, it will be difficult to calculate the pose; [0004] 2. Target recognition at the end of the robotic arm is easily disturbed by ambient light; [0005] 3. In the pose measurement based on monocular vision, it is difficult to match the target feature point with the known feature point model; in the pose measurement based on binocular vision, it is dif...

AI Technical Summary

Problems solved by technology

[0003] 1. Due to the large range of motion at the end of the robotic arm, there is an occlusion problem. When the end of the robotic arm is blocked, it will be difficult to calculate the pose;

[0004] 2. Target recognition at the end of the robotic arm is easily disturbed by ambient light;

[0005] 3. In the pose measurement based on monocular vision, it is difficult to match the target feature point with the known feature point model; in the pose measurement based on binocular vision, it is difficult to match the same feature point of two images

Images