Mechanical arm intelligent teaching method based on computer vision and application

Abstract

The invention relates to a mechanical arm intelligent teaching method based on computer vision and application, and the method comprises the steps: 1, enabling a target object to move along an expected path, and inputting a color image into a target detection model for detection; 2, inputting a detection result into a target tracking model for tracking to obtain a pixel target path; 3, mapping thedepth map to a color map to obtain a target path Path 1; 4, performing three-dimensional reconstruction on the point cloud to obtain a point cloud image of the target path Path1; 5, performing line feature extraction on the point cloud image of the target path Path1 to obtain a line feature set; matching the target path Path 1 with the line feature set to obtain a target path Path 2; and 6, performing hand-eye calibration on the mechanical arm to obtain a target path Path 3, wherein the target path Path 3 is the working path of the mechanical arm. The method can be applied to a building scene, so that the mechanical arm completes welding, assembling or repairing tasks.

Description

technical field [0001] The invention belongs to the technical field of intelligent buildings, in particular to a computer vision-based intelligent teaching method of a manipulator and its application. Background technique [0002] At present, the teaching methods of industrial robots are generally divided into three types: online teaching, offline teaching and traction teaching. [0003] Online teaching is a teaching method adopted by most robots, that is, the engineer manipulates the teaching box to control the movement of the robotic arm and complete the expected actions to obtain the operation trajectory. The advantage of online teaching is that there is no need to establish an environment model offline. The disadvantage is that it is necessary to manually and accurately operate the movement of the robotic arm to obtain an accurate operation trajectory. The teaching process is cumbersome and inefficient. [0004] Offline teaching is to reconstruct the 3D virtual working ...

AI Technical Summary

Problems solved by technology

The disadvantage is that the cost is too high, the load is small, and at the same time, the worker is in direct contact with the powered manipulator during the whole teaching process, so the safety factor is very low

[0006] To sum up, in the face of constantly changing working scenes, changing working positions at any time, and complex architectural scenes, such as grinding walls, welding steel frames, etc., the above three teaching methods cannot efficiently complete the teaching work, so this application proposes A method of intelligent teaching of manipulator based on computer vision

Images