Genetic Algorithm for Calibrating a Three-Axis Measuring System

Abstract

It is very difficult to avoid the orthoerror and gain error in the three-axis measuring device by means of improving the manufacturing process, and the measure precision is tremendously limited by these errors. A model for this study was built by the mapping relationship between the measure system and the virtual orthosystem. A genetic algorithm was proposed for calibrating the orthoerror and gain error among three measuring axes and was compared with other existing approaches presented in the literature. The results of simulation demonstrate that the genetic algorithm shows a better performance than other algorithms, and a triaxial measure system can be calibrated more efficiently by genetic algorithm.