Research on Precision Motion Control of Micro-motion Platform Based on Uncalibrated Visual Servo

Abstract

A visual servo control algorithm based on back propagation (BP) neural network and genetic algorithm is proposed for the problem of slow recognition speed and low accuracy of visual servo control. The algorithm models the robot and image complex Jacobi matrix to get the initial BP neural network visual servo controller, and then uses genetic algorithm to train the initial weights and thresholds of the controller to finally obtain the hybrid optimized visual control model, which can effectively combine the good global search ability of genetic algorithm with the accurate local search function of BP neural network. The experimental results show that the convergence speed is accelerated while the error is reduced to 4.6% of the original one, which provides a simple and effective method for robot control.