基于粒子滤波的机器人主动定位算法  

Active robot localization algorithm based on particle filter

作　　者：常天莉 黄浩晖 陈玮[1] CHANG Tian-li;HUANG Hao-hui;CHEN Wei(School of Automation,Guangdong University of Technology,Guangzhou 510006,China)

机构地区：[1]广东工业大学自动化学院,广东广州510006

出　　处：《计算机工程与设计》2018年第2期570-573,共4页Computer Engineering and Design

摘　　要：为提高机器人主动-全局定位问题中的精度和效率,利用RoboCup足球机器人仿真平台,结合改进的粒子滤波算法,提出一种具有鲁棒性的启发式主动定位算法。该算法的核心思想是利用改进的粒子滤波算法,对粒子簇进行迭代更新;当粒子簇方差较大时,通过改进的启发式函数,结合误差函数优化策略规划来寻找机器人自定位的最优解。仿真结果表明,该算法在2D足球定位问题中能有效提高球员的定位精度以及定位速度。To improve the accuracy and efficiency in the active-global localization problem,a robustness heuristic localization approach using the improved particle filters based on RoboCup soccer simulation platform was presented to improve the position accuracy and increase the solution speed.The core idea of this framework was to use the modified particle filters to update the clusters,and a modified heuristic function and error function were used to optimize the strategy planning in sake of deriving the optimal solution for self-localisation.Extensive simulations show that the proposed algorithm is much effective and efficient for improving the localization accuracy and speed in RoboCup 2D soccer localization problem.

关 键 词：主动定位 粒子滤波器 策略规划 启发式函数 机器人足球 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置] TP391.9[自动化与计算机技术—控制科学与工程]