基于萤火虫算法改进移动机器人定位方法研究  

Research on the improved mobile robot localization approach based on firefly algorithm

作　　者：朱奇光[1,2] 肖亚昆 陈卫东[1,2] 倪春香 陈颖[3] 

机构地区：[1]燕山大学信息科学与工程学院,秦皇岛066004 [2]河北省特种光纤与光纤传感重点实验室,秦皇岛066004 [3]燕山大学电气工程学院,秦皇岛066004

出　　处：《仪器仪表学报》2016年第2期323-329,共7页Chinese Journal of Scientific Instrument

基　　金：国家自然科学基金(61201112);河北省自然科学基金(F2012203169);河北省普通高等学校青年拔尖人才计划(BJ2014056);燕山大学青年教师自主研究计划(14LGA013)项目资助

摘　　要：针对传统蒙特卡洛定位中粒子退化以及粒子贫乏造成的移动机器人定位精度下降问题,提出了利用萤火虫算法改进蒙特卡洛定位的方法。利用改进后的萤火虫算法优化粒子的采样过程,使粒子在权值更新前趋向高似然区域,并且改进了重采样策略,新的重采样可以使粒子的多样性更好。将改进后的新算法用于机器人定位实验中,结果表明新算法相比扩展卡尔曼粒子滤波在粒子数分别为10、30、50的情况下性能分别提高了20%、34%、29%,并且使用的时间更少。Aiming at the problem that the positioning accuracy of the mobile robot decreases due to the degradation and poor of the particles in traditional Monte Carlo localization,a method is proposed,which uses the firefly algorithm to improve the Monte Carlo localization. The improved firefly algorithm is used to optimize the sampling process of the particles,which makes the particles tend to the high likelihood region before the weight update and improves the resampling strategy. The new resampling strategy can make the particle diversity better. The improved new algorithm was used in robot localization experiments and compared with the traditional method,The results show that compared with the extended Kalman particle filtering algorithm,the new algorithm can improve the performance by 20%,34% and 29% when the number of the particles is 10,30 and 50,respectively; the time consumption is less,and the operating efficiency is improved.

关 键 词：移动机器人 蒙特卡洛定位 粒子滤波 萤火虫算法 重采样 

分 类 号：TP89[自动化与计算机技术—检测技术与自动化装置]