无迹粒子滤波在多机器人同时定位与建图中的应用  

Application of unscented particle filter in simultaneous localization and mapping of multiple robots

作　　者：董文彬[1] 张雅晶[1] 方树平 张华[1] DONG Wen-bin;ZHANG Ya-jing;FANG Shu-ping;ZHANG Hua(Anhui Science and Technology University,Bengbu 233000,China)

机构地区：[1]安徽科技学院,安徽蚌埠233000

出　　处：《黑龙江科学》2018年第12期54-55,共2页Heilongjiang Science

基　　金：安徽科技学院引进人才项目"流体动压轴承微型沟槽加工工艺研究"(JXYJ201605);安徽科技学院引进人才项目"汽车车身用高强度钢板的激光弯曲成型机理研究"(JXYJ201606)

摘　　要：近年来,智能移动机器人的应用领域越来越广泛,从室内运行发展到了室外。由于室外环境空间较大,解决多机器人问题就变得越来越重要了。研究了一种用多机器人解决同时定位和建立地图(SLAM)的算法。这种算法是基于无迹粒子滤波(UPF)进行建立的。利用无迹粒子滤波可以获得比扩展卡尔曼滤波(EKF)更好的结果。利用这两种滤波方法对多机器人同时定位和建立地图问题进行仿真,并对比了仿真结果。通过仿真来验证UPF比EKF更加有效。In recent years,the application fields of intelligent mobile robots are more and more extensive,from indoor operation to outdoor. Because of the large outdoor environment,it is becoming more and more important to solve the problem of multi robots. A multi robot algorithm for simultaneous localization and mapping（ SLAM） is studied. The algorithm is based on Unscented Particle Filter（ UPF）. Using the unscented particle filter,we can get better results than the extended Calman filter（ EKF）. These two filtering methods are used to simulate and map the problem of simultaneous localization and establishment of multiple robots,and the simulation results are compared. The simulation is used to verify that UPF is more effective than EKF.

关 键 词：无迹滤波 同时定位与建图 多机器人 卡尔曼滤波 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TP242[自动化与计算机技术—计算机科学与技术]