类GPS超声定位系统中自主导引小车动态定位算法  

A dynamic location algorithm for autonomous guide vehicle in Quasi-GPS ultrasonic location system

作　　者：苏绍景[1] 程晓畅[1] 王跃科[1] 

机构地区：[1]国防科技大学机电工程与自动化学院仪器系,长沙410073

出　　处：《应用声学》2008年第2期155-160,共6页Journal of Applied Acoustics

摘　　要：针对自主导引小车(AGV)常规超声波局部定位系统的缺陷,提出了类GPS超声定位系统并阐述了其工作原理。研究了小车动态情况下与定位基站的距离估计问题,在获得AGV与定位基站的观测距离后可通过Gauss-Newton迭代定位算法获得小车空间位置的粗估计值。仿真结果表明,将上述获得的粗估计值序列经过卡尔曼滤波器进一步处理后,可提高AGV动态定位精度和速度的估计精度。In view of the deficiencies of the conventional ultrasonic location system, a Quasi-GPS ultrasonic location system for Autonomous Guide Vehicle is presented and its working principle is introduced in this article. The dynamic distance estimation problem between the AGV and location base-station is studied. Coarse estimation value of AGV coordinate can be obtained by the Gauss-Newton Iteration location algorithm from the observed distances among the AGV and the location base-stations. It is shown by simulation that location and velocity estimation precision can be augmented if the above coarse estimation values are processed by Kalman filter.

关 键 词：类GPS 自主导引小车 动态定位 短时傅立叶变换 卡尔曼滤波 

分 类 号：TB565[交通运输工程—水声工程] TN929.53[理学—物理]