模型参数未知时的CPHD多目标跟踪方法

doi:10.3969/j.issn.1001-2400.2017.02.007

Abstract

Abstract:

Since the multi-object tracking performance of the traditional method will decline with unknown model parameters, a CPHD target tracking algorithm is proposed to jointly estimate the detection probability and measurement noise covariance. Firstly, for model the unknown parameters of multiple targets tracking, the detection probability is considered as a variable in a distribution. The detection probability can be obtained by estimating the mean of the distribution. Then, the Variational Bayesian method is used to estimate the covariance of the measurement noise. Finally, the Gaussian implementation of this algorithm is presented. Simulation results show that the algorithm has good tracking performance under jointly unknown detection probability and the covariance of the measurement noise.

Key words: detection probability, measurement noise covariance, variational Bayesian, multitarget tracking

LI Cuiyun;WANG Jingyi;JI Hongbing;WANG Rong. CPHD multi-target tracking algorithm with unknown model parameters[J].Journal of Xidian University, 2017, 44(2): 37-41+184.

References

［1］ BEARD M, REUTER S, GRANSTROM K, et al. Multiple Extended Target Tracking with Labeled Random Finite Sets［J］. IEEE Transactions on Signal Processing, 2016, 64(7): 1638-1653.
［2］ YAZDIAN-DEHKORDI M, AZIMIFAR Z. An Improvement on GM-PHD Filter for Target Tracking in Presence of Subsequent Miss-detection［C］//Proceedings of the 2015 23rd Iranian Conference on Electrical Engineering. Piscataway: IEEE, 2015: 765-769.
［3］ LI S, YI W, WANG B, et al. A Generalized Expression of Multi-target Probability Density in the Framework of FISST［C］//Proceedings of the IEEE National Radar Conference. Piscataway: IEEE, 2015: 986-991.
［4］ CLARK D, VO B N, BELL J. GM-PHD Filter Multitarget Tracking in Sonar Images［C］//Proceedings of SPIE: 62350. Bellingham: SPIE, 2006: 62350R.
［5］ LI C Y, WANG R, HU Y H, et al. Cardinalised Probability Hypothesis Density Tracking Algorithm for Extended Objects with Glint Noise［J］. IET Science, Measurement ＆ Technology, 2016, 10(5): 528-536.
［6］ MAHLER R P S. Multitarget Bayes Filtering via First-order Multitarget Moments［J］. IEEE Transactions on Aerospace and Electronic Systems, 2003, 39(4): 1152-1178.
［7］郑学弢. 复杂环境下基于随机集的多目标跟踪算法研究［D］. 西安: 西安电子科技大学, 2013.
［8］ OUSSALAH M, de SCHUTTER J. Adaptive Kalman Flter for Noise Identification［C］//Proceedings of the 25th International Conference on Noise and Vibration Engineering. Leuven: Katholieke Universiteit Leuven, 2000: 1247-1254.
［9］ ZHU Y M. Efficient Recursive State Estimator for Dynamic Systems without Knowledge of Noise Covariances［J］. IEEE Transactions on Aerospace and Electronic Systems, 1999, 35(1): 102-114.
［10］ LI X R, BAR-SHALOM Y. A Recursive Multiple Model Approach to Noise Identification［J］. IEEE Transactions on Aerospace and Electronic Systems, 1994, 30(3): 671-684.
［11］ YANG J L, GE H W. Adaptive Probability Hypothesis Density Filter Based on Variational Bayesian Approximation for Multi-target Tracking［J］. IET Radar, Sonar ＆ Navigation, 2013, 7(9): 959-967.
［12］李翠芸, 江舟, 姬红兵. 一种新的未知杂波环境下的PHD 滤波器［J］. 西安电子科技大学学报, 2014, 41(5): 18-23.
LI Cuiyun, JIANG Zhou, JI Hongbing. A Novel PHD Filter in Unknown Clutter Environment［J］. Journal of Xidian University. 2014, 41(5): 18-23.

[1]	LI Cuiyun;JIANG Zhou;LI Bin;ZHOU Xuan. Gaussian mixture PHD smoothing filter in unknown clutter [J]. J4, 2015, 42(5): 98-104.
[2]	LI Cuiyun;JIANG Zhou;JI Hongbing. Novel PHD filter in unknown clutter environment [J]. J4, 2014, 41(5): 18-23.
[3]	ZHU Yan;CUI Yan-peng;ZHAO Guo-qing. Sinusoidal weighted frequency modulation jamming technique against moving target detection radar [J]. J4, 2010, 37(2): 273-278.
[4]	ZHANG Ying-guang;LIAO Gui-sheng;ZHANG Lin-rang. A new adaptive PN code blind acquisition scheme with CFAR [J]. J4, 2004, 31(6): 855-859.

CPHD multi-target tracking algorithm with unknown model parameters

PDF (PC)

Like

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 4

Metrics

Comments

Recommended 10