基于激光雷达传感器的RBPF-SLAM系统优化设计
|
摘要
针对传统Rao-Blackwellized粒子滤波(RBPF)方法的同步定位和建图(SLAM)问题,提出了一种基于激光雷达的RBPF-SLAM系统优化方法,利用高精度激光雷达数据,修正了基于里程计读数的建议分布函数,减少了滤波过程所需的粒子数目;引入了自适应重采样机制,缓解由于重采样带来的粒子消耗问题。为验证改进算法性能,在搭建的差速型移动机器人平台上,进行了验证试验,结果表明:改进后的RBPF-SLAM方法,能够实时构建栅格地图,在建图效率和精度上均有明显的提升。
Aiming at problem of simultaneous localization and mapping( SLAM) of traditional Rao-Blackwellized particle filtering( RBPF) method,an optimization method of RBPF-SLAM system based on laser radar( LIDAR) is proposed. This method uses high precision LIDAR data to correct the proposed distribution function based on the odometer readings and reduce number of particles required for filtering process. At the same time,the adaptive resampling mechanism is introduced to alleviate the problem of particle consumption due to resampling. In order to verify the performance of the improved algorithm,a method validation test is carried out on the platform of the differential mobile robot. The results show that the improved RBPF-SLAM method can build the gridding map in real time,and it can improve the efficiency and precision of mapping.
Aiming at problem of simultaneous localization and mapping( SLAM) of traditional Rao-Blackwellized particle filtering( RBPF) method,an optimization method of RBPF-SLAM system based on laser radar( LIDAR) is proposed. This method uses high precision LIDAR data to correct the proposed distribution function based on the odometer readings and reduce number of particles required for filtering process. At the same time,the adaptive resampling mechanism is introduced to alleviate the problem of particle consumption due to resampling. In order to verify the performance of the improved algorithm,a method validation test is carried out on the platform of the differential mobile robot. The results show that the improved RBPF-SLAM method can build the gridding map in real time,and it can improve the efficiency and precision of mapping.
引文
[1]Smith R,Self M,Cheeseman P.Estimating uncertain spatial relationships in robotics[C]∥Proceedings of 1987 IEEE International Conference on Robotics and Automation,North-Holland:1987:850.
[2]Hammersley J M,Morton K W.Poorman's Monte Carlo[J].J of the Royal Statistical Society B,1954,16(1):23-38.
[3]Murphy K,Russell S.Rao-Blackwellized particle filtering for dynamic Bayesian networks,sequential Monte Carlo methods[M].Lorne,Australia:Springer-Verlag,2001:499-515.
[4]Montemerlo M,Thrun S,Koller D,et al.Fast SLAM:A factored solutionto the simultaneous localization and mapping problem[C]∥Proceedings of the AAAI National Conference on Artificial Intelligence,Menlo Park,CA,USA:AAA I,2002:593-598.
[5]Montemerlo M,Thrun S,Koller D,et al.Fast SLAM 2.0:An improved particle filtering algorithm for simultaneous localization and mapping that provably converges[C]∥Proceedigns of the Sixteenth International Joint Conference on Artificial Intelligence(IJCAI),Menlo Park,CA,USA:AAAI,2003:1151-1156.
[6]历茂海,洪炳荣,罗荣华.用改进的Rao-blackwellized粒子滤波器实现移动机器人同时定位和地图创建[J].吉林大学学报,2007,37(2):402-406.
[7]祝继华,郑南宁,袁泽剑,等.基于ICP算法和粒子滤波的未知环境地图创建[J].自动化学报,2009,35(8):1107-1113.
[8]Giorgio Grisetti,Cyrill Stachniss,Wolfram Burgard.Improving grid-based SLAM with Rao-Blackwellized particle filters by adaptive proposals and selective resampling[C]∥IEEE International Conference on robotics and Automation(ICRA),2005.
[9]Giorgio Grisetti,Cyrill Stachniss,Wolfram Burgard.Improved techniques for grid mapping with Rao-Blackwellized particle filters[J].IEEE Transactions on Robotics,2007,23:34-46.
[10]Besl P J,Mckay N D.Method for registration of 3D shapes[J].IEEE Transactions on Pattern Analysis&Machine Intelligence,1992,14(2):239-256.
[11]沈林成,徐昕,朱华勇,等.移动机器人自主控制技术[M].北京:科学出版社,2011:248-269.
[12]张建伟,张立伟,胡俊,等.开源机器人操作系统ROS[M].北京:科学出版社,2012.
[13]Bolic M,Djuric P M,Hong S.New resampling algorithms for particle filters[C]∥IEEE International Conference on Acoustics,Speech,and Signal Processing,Hong Kong:IEEE,2003:589-592.
[14]罗元,苏琴,张毅,等.基于优化RBPF的同时定位与地图构建[J].华中科技大学学报:自然科学版,2016,44(5):30-34.
[2]Hammersley J M,Morton K W.Poorman's Monte Carlo[J].J of the Royal Statistical Society B,1954,16(1):23-38.
[3]Murphy K,Russell S.Rao-Blackwellized particle filtering for dynamic Bayesian networks,sequential Monte Carlo methods[M].Lorne,Australia:Springer-Verlag,2001:499-515.
[4]Montemerlo M,Thrun S,Koller D,et al.Fast SLAM:A factored solutionto the simultaneous localization and mapping problem[C]∥Proceedings of the AAAI National Conference on Artificial Intelligence,Menlo Park,CA,USA:AAA I,2002:593-598.
[5]Montemerlo M,Thrun S,Koller D,et al.Fast SLAM 2.0:An improved particle filtering algorithm for simultaneous localization and mapping that provably converges[C]∥Proceedigns of the Sixteenth International Joint Conference on Artificial Intelligence(IJCAI),Menlo Park,CA,USA:AAAI,2003:1151-1156.
[6]历茂海,洪炳荣,罗荣华.用改进的Rao-blackwellized粒子滤波器实现移动机器人同时定位和地图创建[J].吉林大学学报,2007,37(2):402-406.
[7]祝继华,郑南宁,袁泽剑,等.基于ICP算法和粒子滤波的未知环境地图创建[J].自动化学报,2009,35(8):1107-1113.
[8]Giorgio Grisetti,Cyrill Stachniss,Wolfram Burgard.Improving grid-based SLAM with Rao-Blackwellized particle filters by adaptive proposals and selective resampling[C]∥IEEE International Conference on robotics and Automation(ICRA),2005.
[9]Giorgio Grisetti,Cyrill Stachniss,Wolfram Burgard.Improved techniques for grid mapping with Rao-Blackwellized particle filters[J].IEEE Transactions on Robotics,2007,23:34-46.
[10]Besl P J,Mckay N D.Method for registration of 3D shapes[J].IEEE Transactions on Pattern Analysis&Machine Intelligence,1992,14(2):239-256.
[11]沈林成,徐昕,朱华勇,等.移动机器人自主控制技术[M].北京:科学出版社,2011:248-269.
[12]张建伟,张立伟,胡俊,等.开源机器人操作系统ROS[M].北京:科学出版社,2012.
[13]Bolic M,Djuric P M,Hong S.New resampling algorithms for particle filters[C]∥IEEE International Conference on Acoustics,Speech,and Signal Processing,Hong Kong:IEEE,2003:589-592.
[14]罗元,苏琴,张毅,等.基于优化RBPF的同时定位与地图构建[J].华中科技大学学报:自然科学版,2016,44(5):30-34.