Improvement on Robots Positioning Accuracy Based on Genetic Algorithm

Abstract

The paper analyzes the robot link's positioning error sources and builds its error model of geometrical parameters. With the aid of the genetic algorithm (GA) that has the powerful global adaptive probabilistic search ability, 24 parameters of a 6-DOF robot are identified through simulation, which makes the robot's position and orientation accuracy an great improvement. In the process of the robot calibration, stochastic measurement noises are considered. The simulation results show that with GA calibrating the robot is a kind of superior method, even if the robot link's parameters are relative, GA still has search ability to find the optimum solution.