基于位姿图优化的势均衡多伯努利滤波器SLAM方法

doi:10.11959/j.issn.2096-6652.202305

智能科学与技术学报 ›› 2023, Vol. 5 ›› Issue (1): 113-120.doi: 10.11959/j.issn.2096-6652.202305

基于位姿图优化的势均衡多伯努利滤波器SLAM方法

张子菁, 章飞

江苏科技大学海洋学院，江苏镇江 212003

修回日期:2022-12-07 出版日期:2023-03-15 发布日期:2023-03-01
作者简介:张子菁（1995- ），男，江苏科技大学海洋学院硕士生，主要研究方向为移动机器人SLAM方法
章飞（1979- ），男，博士，江苏科技大学副教授，主要研究方向为水下自主导航、MIMO雷达信号处理、多传感器信息融合
基金资助:
国家自然科学基金资助项目(61801170);国家自然科学基金资助项目(61801435)

Cardinalized balanced multi-Bernoulli filter SLAM method based on pose graph optimization

Zijing ZHANG, Fei ZHANG

Ocean College, Jiangsu University of Science and Technology, Zhenjiang 212003, China

Revised:2022-12-07 Online:2023-03-15 Published:2023-03-01
Supported by:
TheNational Natural Science Foundation of China(61801170);TheNational Natural Science Foundation of China(61801435)

摘要/Abstract

摘要：

在复杂室内环境下，传统基于随机有限集理论的SLAM方法存在机器人位姿精度低、计算量大的问题。针对此问题，提出一种基于位姿图优化的势均衡多伯努利滤波器SLAM方法。首先，该方法通过势均衡多伯努利滤波器获得地图特征估计，避免了数据关联。其次，提出了自适应信息控制法，丰富先验信息。然后，通过自适应信息控制法将位姿图优化理论与势均衡多伯努利滤波器SLAM结合，优化机器人的位姿估计。最后，进行实验对比分析，结果表明所提方法比RB-PHD-SLAM方法有更好的SLAM精度及实时性。

关键词: SLAM, 随机有限集理论, 势均衡多伯努利滤波器, 位姿图优化

Abstract:

In the complex indoor environment, the traditional SLAM method based on random finite set theory has the problems of low robot pose accuracy and large amount of calculation.To solve these problems, a cardinalized balanced multi-Bernoulli filter SLAM method based on pose graph optimization was proposed.First of all, the cardinalized balanced multi-Bernoulli filter was used to estimate the map features, which avoided data association.What is more, an adaptive information control method was proposed to enrich the prior information.Then, the pose graph optimization theory was combined with cardinalized balanced multi-Bernoulli filter SLAM through adaptive information control method to optimize the pose estimation of the robot.Finally, through experimental comparative analysis, the results show that this method have better SLAM accuracy and real-time performance than the RB-PHD-SLAM method.

Key words: SLAM, random finite set theory, cardinalized balanced multi-Bernoulli filter, posegraph optimization

中图分类号:

TP242

张子菁,章飞. 基于位姿图优化的势均衡多伯努利滤波器SLAM方法[J]. 智能科学与技术学报, 2023, 5(1): 113-120.

Zijing ZHANG,Fei ZHANG. Cardinalized balanced multi-Bernoulli filter SLAM method based on pose graph optimization[J]. Chinese Journal of Intelligent Science and Technology, 2023, 5(1): 113-120.

图/表 12

图1

图2

表1

表2

图3

图4

图5

图6

图7

图8

图9

表3

参考文献 20

[1]	KEGELEIRS M , GRISETTI G , BIRATTARI M . Swarm SLAM:challenges and perspectives[J]. Frontiers in Robotics and AI, 2021,8.
[2]	李伟, 任孟瀚, 黄威豪 ,等. 基于改进M-ORB的视觉SLAM直接闭环检测算法[J]. 智能科学与技术学报, 2021,3(4): 482-491.
	LI W , REN M H , HUANG W H ,et al. Improved M-ORB based direct-loop closure detection algorithm for visual SLAM[J]. Chinese Journal of Intelligent Science and Technology, 2021,3(4): 482-491.
[3]	FERRERA M , EUDES A , MORAS J ,et al. OV²SLAM:a fully online and versatile visual SLAM for real-time applications[J]. IEEE Robotics and Automation Letters, 2021,6(2): 1399-1406.
[4]	TANG M , CHEN Z , YIN F L . Robot tracking in SLAM with masreliez-martin unscented Kalman filter[J]. International Journal of Control,Automation and Systems, 2020,18(9): 2315-2325.
[5]	HUANG Y W , SHAN T X , CHEN F F ,et al. DiSCo-SLAM:distributed scan context-enabled multi-robot liDAR SLAM with two-stage global-local graph optimization[J]. IEEE Robotics and Automation Letters, 2022,7(2): 1150-1157.
[6]	MAHLER R . Multitarget Bayes filtering via first-order multitarget moments[J]. IEEE Transactions on Aerospace and Electronic Systems, 2003,39(4): 1152-1178.
[7]	MAHLER R . PHD filters of higher order in target number[J]. IEEE Transactions on Aerospace and Electronic Systems, 2007,43(4): 1523-1543.
[8]	MAHLER R . Statistical multisource-multitarget information fusion[M]. Boston: Artech House, 2007: 565-682.
[9]	VO B N , SINGH S , DOUCET A . Sequential Monte Carlo methods for multitarget filtering with random finite sets[J]. IEEE Transactions on Aerospace and Electronic Systems, 2005,41(4): 1224-1245.
[10]	VO B N , MA W K . The Gaussian mixture probability hypothesis density filter[J]. IEEE Transactions on Signal Processing, 2006,54(11): 4091-4104.
[11]	ADAMS M , VO B N , MAHLER R ,et al. SLAM gets a PHD:new concepts in map estimation[J]. IEEE Robotics ＆ Automation Magazine, 2014,21(2): 26-37.
[12]	MORATUWAGE D , ADAMS M , INOSTROZA F . δ-generalized labeled multi-bernoulli simultaneous localization and mapping with an optimal kernel-based particle filtering approach[J]. Sensors, 2019,19(10): 2290.
[13]	DEUSCH H , REUTER S , DIETMAYER K . The labeled multi-bernoulli SLAM filter[J]. IEEE Signal Processing Letters, 2015,22(10): 1561-1565.
[14]	闫德立, 宋永端, 宋宇 ,等. 一种改进的高斯混合概率假设密度SLAM算法[J]. 控制与决策, 2014,29(11): 1959-1965.
	YAN D L , SONG Y D , SONG Y ,et al. An improved Gaussian mixture PHD SLAM algorithm[J]. Control and Decision, 2014,29(11): 1959-1965.
[15]	GRISETTI G , KüMMERLE R , STACHNISS C ,et al. A tutorial on graph-based SLAM[J]. IEEE Intelligent Transportation Systems Magazine, 2010,2(4): 31-43.
[16]	THRUN S , LIU Y F , KOLLER D ,et al. Simultaneous localization and mapping with sparse extended information filters[J]. The International Journal of Robotics Research, 2004,23(7/8): 693-716.
[17]	ZHANG Q Y , LI Y , MA T ,et al. Bathymetric particle filter SLAM with graph-based trajectory update method[J]. IEEE Access, 2021,9: 85464-85475.
[18]	KüMMERLE R , STEDER B , DORNHEGE C ,et al. Large scale graph-based SLAM using aerial images as prior information[J]. Autonomous Robots, 2011,30(1): 25-39.
[19]	SCHUHMACHER D , VO B T , VO B N . A consistent metric for performance evaluation of multi-object filters[J]. IEEE Transactions on Signal Processing, 2008,56(8): 3447-3457.
[20]	CHEVUL S , BINZENHFER A , SCHMID M ,et al. A self-organizing concept for distributed end-to-end quality monitoring[J]. Informatik Uni, 2006.

步骤	内容
步骤1	初始化第0时刻的机器人位姿及地图特征RFS
步骤2	初始化运行时刻数k，优化时刻数t及n_z(X_i)
步骤3	机器人位姿预测，得到在 CBMber-SLAM 中的机器人预测位姿
步骤4	按照式（5）～式（15）进行对应时刻数t的地图特征预测过程和更新过程
步骤5	经过高斯项修剪、合并、地图特征提取，获得地图特征的状态，输出地图特征估计值
步骤6	判断信息控制量值 A 是否小于距离阈值，若小于则执行步骤8，否则执行步骤7
步骤7	将当前时刻获得的地图特征对应的高斯分量加入下一步地图特征估计过程的新生地图集合中，丰富先验信息，然后执行步骤3，令k=k+1，t=t+1
步骤8	若A大于数目阈值，则记录该t个时刻数所对应的机器人位姿、地图特征位姿信息
步骤9	对步骤 3～步骤 8 获得的机器人位姿信息、地图特征信息进行图构建
步骤10	对步骤9中构建完成的图进行优化，获得最优的机器人位姿。并更新该 t 个时刻数所对应的机器人位姿，令 t=0， k=k+1，输出机器位姿估计值，并跳转至步骤3

参数	值
机器人移动速度/(m·s^-1)	10
过程噪声/m	0.3
雷达扫描范围	150°
雷达扫描半径距离/m	20
角度观测噪声	1°
距离观测噪声/m	0.1

方法	机器人位姿误差/m	特征数目	特征数目误差	OSPA误差/m	时间/s
RB-PHD-SLAM方法	5.749	45	27	12.173	0.024
CBMber-SLAM方法	1.284	18	0	3.007	0.021
GB-CBMber-SLAM方法	0.525	18	0	3.004	0.019

基于位姿图优化的势均衡多伯努利滤波器SLAM方法

Cardinalized balanced multi-Bernoulli filter SLAM method based on pose graph optimization

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献 20

相关文章 1

Metrics

推荐阅读 0