基于多目标不确定性改进的GM-PHD滤波器

doi:10.12382/bgxb.2021.0693

摘要/Abstract

摘要： 基于随机有限集的高斯混合概率假设密度（GM-PHD）滤波是处理多目标跟踪问题的一种有效方法。GM-PHD滤波器在密集杂波环境中会因估计误差过大而导致跟踪性能的下降，主要是因为没有充分考虑来自多目标量测的不确定性。为此，提出在考虑高斯分量权重的情况下，通过分量值改变协方差更新式，并通过引入标签，采用自适应阈值对高斯分量进行合并。理论分析和仿真结果表明：该方法在杂波环境下，目标最优次模式分配距离小，跟踪精度更高；目标数量的估计结果受杂波的影响更小，其估计值更接近真实的目标数量；通过具有不同杂波以及检测概率条件的跟踪场景，证明了该方法的目标数量估计精度和滤波性能明显好于传统算法。

关键词: 多目标跟踪, 随机有限集, 高斯混合概率假设密度滤波器, 高斯混合, 状态估计

Abstract: Gaussian mixture probability hypothesis density (GM-PHD) filtering based on random finite set (RFS) is an effective method for solving multi-target tracking (MTT) problem. However, the GM-PHD filter in a dense clutter environment may have a poorer tracking performance due to excessive estimation errors. This is mainly because the uncertainty from the multi-target measurement is not fully considered. Every moment, the GM-PHD filter generates a new Gaussian component and a specific measured value. When clutter density is high or the detection probability is low, the accuracy of the estimated value of the target is lower. To solve this problem, this paper proposes a change in the covariance update formula through adjusting the component value with the weight of the Gaussian component considered, and introduces a label to merge the Gaussian components with an adaptive threshold, so as to improve the accuracy of the target number estimation. The proposed algorithm is proven to have a significantly higher target number estimation accuracy and filtering performance than traditional algorithms through tests under tracking scenarios with different clutter and detection probability conditions.

Key words: multi-targettracking, RFS, GM-PHDfilter, gaussianmixture, stateestimation

中图分类号:

V556
TP391

王奎武, 张秦, 虎小龙. 基于多目标不确定性改进的GM-PHD滤波器[J]. 兵工学报, 2022, 43(12): 3113-3121.

WANG Kuiwu, ZHANG Qin, HU Xiaolong. Improved GM-PHD Filter Based on Multi-target Uncertainty[J]. Acta Armamentarii, 2022, 43(12): 3113-3121.

参考文献

［1］ REID D. An algorithm for tracking multiple targets[J]. IEEE Transaction on Automatics Control, 1979, 24(6):1202-1211.
[2] DAUM F. Multitarget-multisensor tracking: principles and techniques [book review][J]. IEEE Aerospace and Electronic Systems Magazine, 1996, 11(2):41-53.
[3] MAHLER R . Multitarget Bayes filtering via first-order multitarget moments[J]. IEEE Transactions on Aerospace & Electronic Systems, 2004, 39(4):1152-1178.
[4] HU Q, JI H B, ZHANG Y Q. A standard PHD filter for joint tracking and classification of maneuvering extended targets using random matrix[J]. Signal Processing, 2018, 144: 352-363.
[5] LEONARD M R, ZOUBIR A M. Multi-target tracking in distributed sensor networks using particle PHD filters[J]. Signal Processing, 2019, 159: 130-146.
[6] LIU Z, JI L N, YANG F B, et al. Cubature information Gaussian mixture probability hypothesis density approach for multi extended target tracking[J]. IEEE Access, 2019, 7: 103678-103692.
[7] FENG P M, WANG W W, DLAY S, et al. Social force model-based MCMC-OCSVM particle PHD filter for multiple human tracking[J]. IEEE Transactions on Multimedia, 2016, 19(4): 725-739.
[8] KIM H, GRANSTRM K, GAO L, et al. Joint CKF-PHD filter and map fusion for 5G multi-cell SLAM[C]∥Proceedings of ICC 2020—2020 IEEE International Conference on Communications. Dublin, Ireland: IEEE, 2020: 1-6.
[9] 姜琦, 王锐, 周超, 等. 基于代数图论的修正贝叶斯群目标航迹起始算法[J]. 电子与信息学报, 2021, 43(3):531-538.
JIANG Q, WANG R, ZHOU C, et al.Modified Bayesian group target track initiation algorithm based on algebraic graph theory[J]. Journal of Electronics and Information Technology, 2021, 43(3): 531-538.(in Chinese)
[10] PANTA K, CLARK D E, VO B N. Data association and track management for the gaussian mixture probability hypothesis density filter[J]. IEEE Transactions on Aerospace & Electronic Systems IEEE, 2009, 45(3):1003-1016.
[11] 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.
[12] LI X R,BAR-SHALOM Y. Tracking in clutter with nearest neighbor filters: analysis and performance[J]. IEEE Transactions on Aerospace and Electronic Systems, 1996, 32(3): 995-1010.
[13] LI X R. Tracking in clutter with strongest neighbor measurements. I. theoretical analysis[J]. IEEE Transactions on Automatic Control, 1998, 43(11): 1560-1578.
[14] BAR-SHALOM Y, KIRUBARAJAN T, LIN X. Probabilistic data association techniques for target tracking with applications to sonar, radar and EO sensors[J]. IEEE Aerospace and Electronic Systems Magazine, 2005, 20(8): 37-56.
[15] YAZDIAN-DEHKORDI M, AZIMIFAR Z, MASNADI-SHIRAZI M A. Penalized Gaussian mixture probability hypothesis density filter for multiple target tracking[J]. Signal Processing, 2012, 92(5):1230-1242.
[16] YAZDIAN-DEHKORDI M, AZIMIFAR Z. Refined GM-PHD tracker for tracking targets in possible subsequent missed detections[J]. Signal Processing, 2015, 116: 112-126.
[17] VO B T, VO B N, CANTONI A. Analytic implementations of the cardinalized probability hypothesis density filter[J]. IEEE transactions on signal processing, 2007, 55(7): 3553-3567.
[18] 杨峰, 王永齐, 梁彦,等. 基于概率假设密度滤波方法的多目标跟踪技术综述[J]. 自动化学报, 2013, 39(11):1944-1956.
YANG F, WANG Y Q, LIANG Y, et al. Overview of multi-target tracking technology based on probability hypothesis density filtering method[J]. Acta Automatica Sinica, 2013, 39(11):1944-1956. (in Chinese)
[19] SONG T L, LEE D G. A probabilistic nearest neighbor filter algorithm for m validated measurements[J]. IEEE Transactions on Signal Processing, 2006, 54(7): 2797-2802.
[20] MAHLER R. PHD filters of higher order in target number[J]. IEEE Transactions on Aerospace and Electronic systems, 2007, 43(4): 1523-1543.

[1]	郑磊, 陈志敏, 贾宇轩. 基于广域部署智能反射面的无人机集群跟踪方法[J]. 兵工学报, 2023, 44(6): 1837-1845.
[2]	任济寰, 吴祥, 薄煜明, 吴盘龙, 何山. 基于增强上下文信息长短期记忆网络的弹道轨迹预测[J]. 兵工学报, 2023, 44(2): 462-471.
[3]	李泳耀, 江磊, 刘宇飞, 孙泽源, 郑冬冬. 仿人灵巧手的稳定抓取方法研究综述[J]. 兵工学报, 2023, 44(11): 3237-3252.
[4]	宋水舟，任会兰，宁建国. 混合型加载下钢纤维混凝土损伤过程的声发射参数分析[J]. 兵工学报, 2022, 43(8): 1881-1891.
[5]	王振峰，李飞，王新宇，杨建森，秦也辰. 基于交互式多模型无迹卡尔曼滤波的悬架系统状态估计[J]. 兵工学报, 2021, 42(2): 242-253.
[6]	邵鹏飞，王蕾，王方勇. 基于序贯蒙特卡洛与概率假设密度滤波的主动分布式声纳多目标跟踪[J]. 兵工学报, 2020, 41(5): 941-949.
[7]	王博洋，龚建伟，熊光明，张瑞增，陈慧岩，席军强. 有级转向履带车辆的驾驶员操控行为模型[J]. 兵工学报, 2020, 41(12): 2379-2388.
[8]	梁文利，陈慧岩，王博洋. 基于独立电驱动履带车辆的地面参量估计方法研究[J]. 兵工学报, 2019, 40(6): 1146-1153.
[9]	卫恒，吕强，刘扬，林辉灿，梁冰. 基于状态切换的分布式多机器人编队控制[J]. 兵工学报, 2019, 40(5): 1103-1112.
[10]	井沛良，段宇，韩超，郭荣化，宁小磊，刘瑜. 基于高斯混合模型和期望最大化算法的非高斯分布圆概率误差估计方法研究[J]. 兵工学报, 2019, 40(2): 369-376.
[11]	王博洋，龚建伟，高天云，陈慧岩，席军强. 基于高斯混合-隐马尔可夫模型的速差转向履带车辆横向控制驾驶员模型[J]. 兵工学报, 2017, 38(12): 2301-2308.
[12]	魏帅, 冯新喜, 王泉, 鹿传国. 基于箱粒子滤波的鲁棒标签多伯努利跟踪算法[J]. 兵工学报, 2017, 38(10): 2062-2068.
[13]	王宏健，徐金龙，李娟，张爱华. 非平稳非高斯测量噪声条件下改进差分粒子滤波算法研究[J]. 兵工学报, 2014, 35(7): 1032-1039.