The Method of Positioning Error Measuring Compensation of the Multiple-Target Motion Simulation System Based on Neural Network
In order to solve the problem of large range of target motion and targets mutual interference in Two-dimensional Cross-Moving Motion Simulation System with Multiple Objects,a multi-objective system is designed,which consists of a number of independent two-dimensional motion simulation systems,and an error measuring compensation program is proposed,which can meet the accuracy requirements.Three-dimensional position coordinates of the space measured feature points are got by two theodolites which is based on triangular intersectional measuring method and the experiment is designed by the best measurement arca which is obtained through the uncertainty analysis.The position accuracy of the measured feature points can be determined by simulation.The movement reference coordinate system can be set up according to the three-dimensional position data of the measured feature points and the spatial position error can be calculated by coordinate system transformation.RBF neural network is used to establish the error compensation model which is finally validated.Experimental results of actual system indicate that the program can be well applied to the Two-dimensional Motion Simulation System for spatial position error measurement and compensation.The accuracy requirements of the system can be satisfied after compensation.
multiple objects two-dimensional cross-moving motion triangle intersectional measurement RBF neural network spatial position error compensation
Huo Ju Dong Wenbo Liu Yunhe
Electrical Engineering Department Harbin Institute of Technology Harbin,China Control and Simulation Center Harbin Institute of Technology Harbin,China
国际会议
杭州
英文
647-652
2013-03-22（万方平台首次上网日期，不代表论文的发表时间）