Design of Industrial Robot Teaching System Based on Machine Vision

Abstract

In this paper, an industrial robot teaching system based on machine vision is proposed using the visual teaching tracking algorithm and the robot’s speed planning algorithm. For traditional industrial robots, it requires professionals to perform programming and teaching, may be clumsy and time-consuming, and cannot meet the needs of enterprises for flexible production. Therefore, the proposed approach in this paper can solve the above problems. Firstly, an optimized moving target trajectory is obtained, which is a MeanShift tracking algorithm based on Kalman filter. Then, combining the S-curve acceleration and deceleration planning algorithm is to control the movement of the robot to smoothly complete the task of trajectory reproduction. Finally, simulations and experiments are presented to illustrate that the teaching system can complete the teaching task quickly, shorten the teaching work time, and has certain significance for expanding the application field of the robot system.