标签箱粒子概率假设密度群目标跟踪算法
|
摘要
针对现有的箱粒子概率假设密度(probability hypothesis density,PHD)群目标跟踪算法计算量大、在群数目较多时状态提取不稳定以及无法获得群的航迹等问题,提出标签箱粒子PHD群目标跟踪算法。该算法首先对量测进行预处理,剔除其中的杂波量测,以降低量测更新的计算量。然后,通过为箱粒子添加标签,区分不同的群目标,获得不同群的航迹。最后,依据不同标签提取群目标的状态,有效避免k-means聚类不稳定带来的影响。仿真实验表明,所提算法具有运算量小,在漏检环境下仍能很好地维持不同群的航迹,并在群数目较多时可准确提取群目标状态等优点。
Aiming at the problems of the existing box particle probability hypothesis density(PHD)based group target tracking algorithm such as a heavy computational burden,poor stability of the extracting state with a large number of clusters and unavailability of group trajectories,a labeled box particle PHD group tracking algorithm is proposed.First,the measurements are preprocessed to eliminate the clutter measurements,so as to reduce the computational burden of the measurement updating step.Then,by adding labels to the box particles,different group targets are distinguished,and the trajectories of different group targets can be obtained.Finally,the states of group targets are extracted according to different labels and the impact of k-means clustering instability are effectively avoided.Simulation experiments illustrate the advantages of the proposed algorithm in terms of light computational burden,track maintenance of different groups under the environment of miss detection,and accurate extraction of the group target state in the case of a large number of clusters.
Aiming at the problems of the existing box particle probability hypothesis density(PHD)based group target tracking algorithm such as a heavy computational burden,poor stability of the extracting state with a large number of clusters and unavailability of group trajectories,a labeled box particle PHD group tracking algorithm is proposed.First,the measurements are preprocessed to eliminate the clutter measurements,so as to reduce the computational burden of the measurement updating step.Then,by adding labels to the box particles,different group targets are distinguished,and the trajectories of different group targets can be obtained.Finally,the states of group targets are extracted according to different labels and the impact of k-means clustering instability are effectively avoided.Simulation experiments illustrate the advantages of the proposed algorithm in terms of light computational burden,track maintenance of different groups under the environment of miss detection,and accurate extraction of the group target state in the case of a large number of clusters.
引文
[1]MIHAYLOVA L,CARMI A Y,SEPTIER F.Overview of Bayesian sequential Mote Carlo methods for group and extended object tracking[J].Digital Signal Processing,2014,25(1):1-16.
[2]KOCH J W.Bayesian approach to extended object and cluster tracking using random matrices[J].IEEE Trans.on Aerospace and Electronic Systems,2008,44(3):1042-1059.
[3]CLARK D,GODSILL S.Group target tracking with the Gaussian mixture probability hypothesis density filter[C]∥Proc.of the3rd International Conference on Intelligent Sensors,Sensor Networks and Information,2007:149-154.
[4]GNING A,MIHAYLOVA L,MASKELL S,et al.Group object structure and state estimation with evolving networks and Monte Carlo methods[J].IEEE Trans.on Signal Processing,2011,59(4):1383-1395.
[5]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.
[6]ABDALLAH F,GNING A,BONNIFAIT P.Box particle filtering for nonlinear state estimation using interval analysis[J].Automatica,2008,44(3):807-815.
[7]ZHU S J,LIU W F,WEN C L,et al.Multiple group targets tracking using the generalized labeled multi-Bernoulli filter[C]∥Proc.of the 35th Chinese Control Conference,2016:4871-4876.
[8]VO B T,VO B N.Labeled random finite sets and multi-object conjugate priors[J].IEEE Trans.on Signal Processing,2013,61(13):3460-3475.
[9]VO B N,VO B T,PHUNG D.Labeled random finite sets and the Bayes multi-target tracking filter[J].IEEE Trans.on Signal Processing,2014,62(24):6554-6567.
[10]宋骊平,刘宇航,程轩.箱粒子PHD演化网络群目标跟踪算法[J].控制与决策,2018,33(1):74-80.SONG L P,LIU Y H,CHENG X.Box-particle evolution network PHD filter for group targets tracking[J].Control and Decision,2018,33(1):74-80.
[11]XUAN C,LIPING S,ZHIBIN Z.Multiple group target tracking with evolving networks and labeled box particle PHD filter[C]∥Proc.of the 30th Chinese Control and Decision Conference,2018:4046-4051.
[12]连峰,韩崇昭,刘伟峰,等.基于SMC-PHDF的部分可分辨的群目标跟踪算法[J].自动化学报,2010,36(5):732-741.LIAN F,HAN C Z,LIU W F,et al.Tracking partly resolvable group targets using SMC-PHDF[J].Acta Automatica Sinica,2010,36(5):732-741.
[13]张慧,徐晖,王雪莹,等.一种基于椭圆随机超曲面模型的高斯混合PHD滤波器[J].光学学报,2013,33(9):6-15.ZHANG H,XU H,WANG X Y,et al.A Gaussian mixture PHD filter for group targets tracking based on ellipse random hypersurface models[J].Acta Optica Sinica,2013,33(9):6-15.
[14]BAUM M,NOACK B,HANEBECK U D.Extended object and group tracking with elliptic random hypersurface models[C]∥Proc.of the 13th Conference on Information Fusion,2010:1-8.
[15]宋骊平,严超,姬红兵,等.基于箱粒子的多扩展目标PHD滤波[J].控制与决策,2015,30(10):1759-1765.SONG L P,YAN C,JI H B,et al.PHD filter for tracking multiple extended targets using box particle[J].Control and Decision,2015,30(10):1759-1765.
[16]朱书军.基于随机有限集的可分辨群目标跟踪算法研究[D].杭州:杭州电子科技大学,2017:29-30.ZHU S J.Multiple resolvable groups tracking by using random finite set[D].Hangzhou:Hangzhou Dianzi University,2017:29-30.
[17]YONGQUAN Z,HONGBING J,QI H.A box-particle implementation of standard PHD filter for extended target tracking[J].Information Fusion,2017,34(1):55-69.
[18]ALLAN D F,LYUDMILA M,AMADOU G,et al.Autonomous crowds tracking with box particle filtering and convolution particle filtering[J].Automatica,2016,69(1):380-394.
[19]SCHIKORA M,AMADOUL G,MIHAYLOVA L,et al.Boxparticle probability hypothesis density filtering[J].IEEE Trans.on Aerospace and Electronic Systems,2014,50(3):1660-1672.
[20]SCHUHMACHER D,VO B T,VO B N.A consistent metric for performance evaluation of multi-object filters[J].IEEETrans.on Signal Processing,2008,56(8):3447-3457.
[2]KOCH J W.Bayesian approach to extended object and cluster tracking using random matrices[J].IEEE Trans.on Aerospace and Electronic Systems,2008,44(3):1042-1059.
[3]CLARK D,GODSILL S.Group target tracking with the Gaussian mixture probability hypothesis density filter[C]∥Proc.of the3rd International Conference on Intelligent Sensors,Sensor Networks and Information,2007:149-154.
[4]GNING A,MIHAYLOVA L,MASKELL S,et al.Group object structure and state estimation with evolving networks and Monte Carlo methods[J].IEEE Trans.on Signal Processing,2011,59(4):1383-1395.
[5]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.
[6]ABDALLAH F,GNING A,BONNIFAIT P.Box particle filtering for nonlinear state estimation using interval analysis[J].Automatica,2008,44(3):807-815.
[7]ZHU S J,LIU W F,WEN C L,et al.Multiple group targets tracking using the generalized labeled multi-Bernoulli filter[C]∥Proc.of the 35th Chinese Control Conference,2016:4871-4876.
[8]VO B T,VO B N.Labeled random finite sets and multi-object conjugate priors[J].IEEE Trans.on Signal Processing,2013,61(13):3460-3475.
[9]VO B N,VO B T,PHUNG D.Labeled random finite sets and the Bayes multi-target tracking filter[J].IEEE Trans.on Signal Processing,2014,62(24):6554-6567.
[10]宋骊平,刘宇航,程轩.箱粒子PHD演化网络群目标跟踪算法[J].控制与决策,2018,33(1):74-80.SONG L P,LIU Y H,CHENG X.Box-particle evolution network PHD filter for group targets tracking[J].Control and Decision,2018,33(1):74-80.
[11]XUAN C,LIPING S,ZHIBIN Z.Multiple group target tracking with evolving networks and labeled box particle PHD filter[C]∥Proc.of the 30th Chinese Control and Decision Conference,2018:4046-4051.
[12]连峰,韩崇昭,刘伟峰,等.基于SMC-PHDF的部分可分辨的群目标跟踪算法[J].自动化学报,2010,36(5):732-741.LIAN F,HAN C Z,LIU W F,et al.Tracking partly resolvable group targets using SMC-PHDF[J].Acta Automatica Sinica,2010,36(5):732-741.
[13]张慧,徐晖,王雪莹,等.一种基于椭圆随机超曲面模型的高斯混合PHD滤波器[J].光学学报,2013,33(9):6-15.ZHANG H,XU H,WANG X Y,et al.A Gaussian mixture PHD filter for group targets tracking based on ellipse random hypersurface models[J].Acta Optica Sinica,2013,33(9):6-15.
[14]BAUM M,NOACK B,HANEBECK U D.Extended object and group tracking with elliptic random hypersurface models[C]∥Proc.of the 13th Conference on Information Fusion,2010:1-8.
[15]宋骊平,严超,姬红兵,等.基于箱粒子的多扩展目标PHD滤波[J].控制与决策,2015,30(10):1759-1765.SONG L P,YAN C,JI H B,et al.PHD filter for tracking multiple extended targets using box particle[J].Control and Decision,2015,30(10):1759-1765.
[16]朱书军.基于随机有限集的可分辨群目标跟踪算法研究[D].杭州:杭州电子科技大学,2017:29-30.ZHU S J.Multiple resolvable groups tracking by using random finite set[D].Hangzhou:Hangzhou Dianzi University,2017:29-30.
[17]YONGQUAN Z,HONGBING J,QI H.A box-particle implementation of standard PHD filter for extended target tracking[J].Information Fusion,2017,34(1):55-69.
[18]ALLAN D F,LYUDMILA M,AMADOU G,et al.Autonomous crowds tracking with box particle filtering and convolution particle filtering[J].Automatica,2016,69(1):380-394.
[19]SCHIKORA M,AMADOUL G,MIHAYLOVA L,et al.Boxparticle probability hypothesis density filtering[J].IEEE Trans.on Aerospace and Electronic Systems,2014,50(3):1660-1672.
[20]SCHUHMACHER D,VO B T,VO B N.A consistent metric for performance evaluation of multi-object filters[J].IEEETrans.on Signal Processing,2008,56(8):3447-3457.