Analysis of Kinematic Parameter Identification Method Based on Genetic Algorithm

For the kinematic parameters calibration of manipulator, different identification methods affect the final calibration results. Comprehensive range error model, the author of this paper, relative distance error model, comprehensive posture error model and relative posture error model of four kinds of error identification model, combined with multiple population genetic algorithm analysis of different methods of parameter identification results, and finally the effect of mechanical arm end position error, and starting from the actual measurement error, to bring the end of the error posture as identification data set, get different adaptive identification method of error, for actual kinematic parameters provide different selection of recognition method based on calibration.