基于Dirichlet过程非参贝叶斯学习的高斯箱粒子滤波快速SLAM算法

doi:10.13973/j.cnki.robot.180580

摘要
图/表
参考文献
相关文章 (14)

全文: PDF (2709 KB) HTML ( ) 输出: BibTeX \| EndNote (RIS)
引用本文:
罗景文, 秦世引. 基于Dirichlet过程非参贝叶斯学习的高斯箱粒子滤波快速SLAM算法[J]. 机器人, 2019, 41(5): 660-675.DOI: 10.13973/j.cnki.robot.180580. LUO Jingwen, QIN Shiyin. Fast SLAM Algorithm Based on the Non-parametric Bayesian Learning ofDirichlet Process for Gauss Box Particle Filtering. ROBOT, 2019, 41(5): 660-675. DOI: 10.13973/j.cnki.robot.180580.

摘要针对常规FastSLAM算法需要大量粒子创建地图以及粒子退化而导致计算复杂度高、难以提高估计精度等问题，提出了一种基于Dirichlet过程非参贝叶斯学习的高斯箱粒子滤波快速SLAM（同步定位与地图构建）算法．首先，改进了箱粒子滤波中以箱粒子为支撑集的均匀概率密度函数，采用高斯概率密度函数进行贝叶斯滤波，提高了估计的精度．在此基础上将Dirichlet过程非参贝叶斯学习应用于高斯箱粒子的重采样，既保证了有效箱粒子数，又能让箱粒子集中在高似然区域，降低了采样枯竭的影响．然后，利用基于Dirichlet过程非参贝叶斯学习的高斯箱粒子滤波进行机器人位姿估计，可有效降低地图创建所需的粒子数，并提高定位精度和实时性．进而采用无迹卡尔曼滤波更新地图特征，以提高地图创建的一致性．仿真结果和轮腿复合机器人实地实验结果验证了本文方法的可行性和有效性．

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	罗景文
	秦世引

关键词 ：同步定位与地图构建, 高斯箱粒子滤波, 非参贝叶斯学习, Dirichlet过程, 无迹卡尔曼滤波

Abstract：The traditional FastSLAM algorithm requires a large number of particles to build maps, and the particle degeneracy will lead to high computational complexity and difficulties in improving the estimation accuracy. For those problems, a fast SLAM (simultaneous localization and mapping) algorithm based on non-parametric Bayesian learning of Dirichlet process for Gauss box particle filter is proposed. Firstly, the uniform probability density function with box particles as the support set in box particle filter is improved, and the Gauss probability density function is used to carry out Bayesian filtering in order to improve the estimation accuracy. On this basis, non-parametric Bayesian learning of Dirichlet process is applied to resampling Gauss box particles, which not only ensures an effective number of box particles, but also enables the box particles to concentrate in the high likelihood region, and thus weakening the effect of sample exhaustion. Then, the non-parametric Bayesian learning of Dirichlet process for Gauss box particle filter is used to estimate the robot pose, which reduces the number of particles needed to build maps and improves the location accuracy and real-time performance. Further, the unscented Kalman filter is used to update the features in the map, which improves the consistency of map building. Simulation and experiment results on the wheel-legged robot verify the feasibility and effectiveness of the proposed method.

Key words： simultaneous localization and mapping Gauss box particle filter non-parametric Bayesian learning Dirichlet process unscented Kalman filter

收稿日期: 2018-09-26 录用日期: 2018-12-19

TP242.6

基金资助:北京市科技计划（D16110400130000-D161100001316001）；国家自然科学基金（61731001，U1435220）

通讯作者: 罗景文,by1503117@buaa.edu.cn;秦世引,qsy@buaa.edu.cn E-mail: by1503117@buaa.edu.cn;qsy@buaa.edu.cn

作者简介: 罗景文(1983-),男,博士生,讲师.研究领域:模式识别与人工智能,机器人同步定位与地图构建.
秦世引(1955-),男,博士,教授,博士生导师.研究领域:控制科学与工程,模式识别与人工智能,智能控制理论及其应用,复杂系统与复杂性科学.

链接本文:

https://robot.sia.cn/CN/10.13973/j.cnki.robot.180580 或 https://robot.sia.cn/CN/Y2019/V41/I5/660

[1] Smith R C, Cheeseman P. On the representation and estimation of spatial uncertainty[J]. International Journal of Robotics Research, 1986, 5(4):56-68.
[2] Martinez-Cantin R, Castellanos J A. Unscented SLAM for largescale outdoor environments[C]//IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway, USA:IEEE, 2005:328-333.
[3] 康轶非,宋永端,宋宇,等.平方根容积卡尔曼滤波在移动机器人SLAM中的应用[J].机器人,2013,35(2):186-193.Kang Y F, Song Y D, Song Y, et al. Square-root cubature Kalman filter and its application to SLAM of an mobile robot[J]. Robot, 2013, 35(2):186-193.
[4] Doucet A, de Freitas N, Murphy K, et al. Rao-Blackwellised particle filtering for dynamic Bayesian networks[C]//16th Conference on Uncertainty in Artificial Intelligence. San Francisco, USA:Morgan Kaufmann Publishers Inc., 2000:176-183.
[5] Anki?han H, Ari F, Tartan E Ö, et al. Square root central difference-based FastSLAM approach improved by differential evolution[J]. Turkish Journal of Electrical Engineering & Computer Sciences, 2016, 24(3):994-1013.
[6] 周武,赵春霞.一种基于遗传算法的FastSLAM2.0算法[J].机器人,2009,31(1):25-32.Zhou W, Zhao C X. A FastSLAM 2.0 algorithm based on genetic algorithm[J]. Robot, 2009, 31(1):25-32.
[7] 朱奇光,袁梅,王梓巍,等.机器人球面单径容积FastSLAM算法[J].机器人,2015,37(6):708-717.Zhu Q G, Yuan M, Wang Z W, et al. A robot spherical simplex-radial cubature FastSLAM algorithm[J]. Robot, 2015, 37(6):708-717.
[8] Castellanos J A, Tardos J D, Schmidt G. Building a global map of the environment of a mobile robot:The importance of correlations[C]//IEEE International Conference on Robotics and Automation. Piscataway, USA:IEEE, 1997:1053-1059.
[9] Bailey T, Nieto J, Nebot E. Consistency of the FastSLAM algorithm[C]//IEEE International Conference on Robotics and Automation. Piscataway, USA:IEEE, 2006:424-429.
[10] Vincke B, Lambert A, Elouardi A. Guaranteed simultaneous localization and mapping algorithm using interval analysis[C]//13th International Conference on Control, Automation, Robotics & Vision. Piscataway, USA:IEEE, 2014:1409-1414.
[11] Le Bars F, Bertholom A, Jan S, et al. Interval SLAM for underwater robots:A new experiment[J]. IFAC Proceedings Volumes, 2010, 43(14):42-47.
[12] Mustafa M, Stancu A, Delanoue N, et al. Guaranteed SLAM-An interval approach[J]. Robotics and Autonomous Systems, 2018, 100:160-170.
[13] Mourad F, Snoussi H, Abdallah F, et al. Guaranteed boxed localization in MANETs by interval analysis and constraints propagation techniques[C]//IEEE Global Telecommunications Conference. Piscataway, USA:IEEE, 2008:415-419.
[14] Abdallah F, Gning A, Bonnifait P. Box particle filtering for nonlinear state estimation using interval analysis[J]. Automatica, 2008, 44(3):807-815.
[15] 李振兴,刘进忙,李松,等.基于箱式粒子滤波的群目标跟踪算法[J].自动化学报,2015,41(4):785-798.Li Z X, Liu J M, Li S, et al. Group targets tracking algorithm based on box particle filter[J]. Acta Automatica Sinica, 2015, 41(4):785-798.
[16] 赵雪刚,宋骊平,姬红兵.量化量测条件下的交互多模型箱粒子滤波[J].西安电子科技大学学报,2014,41(6):37-44.Zhao X G, Song L P, Ji H B. Interacting multiple model box particle filter with quantitative measurements[J]. Journal of XiDian University, 2014, 41(6):37-44.
[17] Gning A, Ristic B, Mihaylova L. Bernoulli particle/box-particle filters for detection and tracking in the presence of triple measurement uncertainty[J]. IEEE Transactions on Signal Processing, 2012, 60(5):2138-2151.
[18] 周建英,王飞跃,曾大军.分层Dirichlet过程及其应用综述[J].自动化学报,2011,37(4):389-407.Zhou J Y, Wang F Y, Zeng D J. Hierarchical Dirichlet processesand their applications:A survey[J]. Acta Automatica Sinica, 2011, 37(4):389-407.
[19] Moore R E, Kearfott B, Cloud M J. Introduction to interval analysis[M]. 1st ed. Philadelphia, USA:SIAM, 2009.
[20] Gning A, Mihaylova L, Abdallah F. Mixture of uniform probability density functions for non linear state estimation using interval analysis[C]//13th International Conference on Information Fusion. 2010. DOI:10.1109/ICIF.2010.5712085.
[21] 于洁,刘昌云,李志汇.箱粒子滤波理论综述[J].电光与控制,2015,22(11):56-60.Yu J, Liu C Y, Li Z H. A survey of box particle filter theory[J]. Electronics Optics & Control, 2015, 22(11):56-60.
[22] Jaulin L, Kieffer M, Didrit O, et al. Applied interval analysis:With examples in parameter and state estimation, robust control and robotics[M]. London, UK:Springer, 2001.
[23] Gning A, Mihaylova L, Abdallah F, et al. Particle filteringcombined with interval methods for tracking applications[M]//Integrated Tracking, Classification, and Sensor Management:Theory and Applications. Hoboken, USA:John Wiley & Sons, Inc., 2016:43-74.
[24] Jaulin L, Walter E. Set inversion via interval analysis for nonlinear bounded-error estimation[J]. Automatica, 1993, 29(4):1053-1064.
[25] 林青,尹建君,张建秋,等.非线性非高斯模型的高斯和滤波算法[J].系统工程与电子技术,2010,32(12):2493-2499.Lin Q, Yin J J, Zhang J Q, et al. Gaussian sum filtering methodsfor nonlinear non-Gaussian models[J]. Systems Engineering and Electronics, 2010, 32(12):2493-2499.
[26] Escobar M D, West M. Bayesian density-estimation and inference using mixtures[J]. Journal of the American Statistical Association, 1995, 90(430):577-588.
[27] Kim C, Sakthivel R, Chung W K. Unscented FastSLAM:A robust and efficient solution to the SLAM problem[J]. IEEE Transactions on Robotics, 2008, 24(4):808-820.
[28] Luo J W, Qin S Y. A fast algorithm of simultaneous localization and mapping for mobile robot based on ball particle filter[J]. IEEE Access, 2018, 6:20412-20429.
[29] Bailey T. SLAM simulations[DB/OL]. (2008-06-10)[2018-08-28]. http://www.personal.acfr.usyd.edu.au/tbailey/software/index.html.
[30] Chen B F, Cai Z X, Zou Z R. A hybrid data association approach for mobile robot SLAM[C]//International Conference on Control, Automation and Systems. Piscataway, USA:IEEE, 2010:1900-1903.
[31] Nebot E. Victoria park dataset[DB/OL]. (2018-05-16)[2018-08-28]. http://www-personal.acfr.usyd.edu.au/nebot/victoriapark.html.
[32] Kerl C, Sturm J, Cremers D. Dense visual SLAM for RGB-D cameras[C]//IEEE/RSJ International Conference on Intelligent Robots and Systems. Piscataway, USA:IEEE, 2013:2100-2106.
[33] Rublee E, Rabaud V, Konolige K, et al. ORB:An efficient alternative to SIFT or SURF[C]//IEEE International Conference on Computer Vision. Piscataway, USA:IEEE, 2011:2564-2571.
[34] 高翔,张涛.视觉SLAM十四讲:从理论到实践[M].1版.北京:电子工业出版社,2017:167-175.Gao X, Zhang T. Visual SLAM fourteen lecture notes:From theory to practice[M]. 1st ed. Beijing:Publishing House of Electronics Industry, 2017:167-175.
[35] Hornung A, Wurm K M, Bennewitz M, et al. OctoMap:An efficient probabilistic 3D mapping framework based on octrees[J]. Autonomous Robots, 2013, 34(3):189-206.