An improved particle filter based on soft computing with application in target tracking

Document Type : Persian Original Article

Author

Faculty of Electrical and Computer Engineering, University of Birjand, Birjand, Iran.

Abstract

Particle filter is one of the most important filters for estimating nonlinear/non-Gussian systems that is used in many applications. In a standard particle filter, since the common state density function of the state is approximated by recursive importance sampling, the dimension of the joint posterior grows with each time step. This causes the algorithm to be rapidly degenerated. Therefore, the use of a resampling strategy is required for guaranteeing a logical approximation of the density function on the entire path. However, in practice, resampling step is performed on marginal space. Since the system may not exhibit an exponential forgetting behavior from its past errors, it will produce an incomplete estimate with a small number of finite sampling processes on the marginal space. To solve this problem, an improved particle filter based soft computing is proposed in this paper. Unlike a particle filter, this filter is applied to the marginal distribution, and the sampling dimensions do not increase with time. In addition, sampling has been improved using an evolutionary differential algorithm. The proposed method is evaluated using computer simulations. The results show that the proposed method has a better performance than standard particle filter.

Keywords

References

[1]        S.Arulampalam,  S.Maskell , N.Gordon, and T.Clapp, “ A tutorial on particle filters for Online onlinear/non-Gaussian Bayesian tracking”, IEEE Trans. Signal Process., 2002, vol. 50, no. 2, pp. 174–188.
[2]        B.Ristic, S.Arulampalam,and N.Gordon, “in Beyond Kalman Filter: Particle filters tracking applications” , 1st ed. Boston, 2004.
[3]        D.Simon, “Optimal State Estimation Kalman, H∞ and Nonlinear Approaches”, John Wiley and Sons, Inc, 2006.
[4]        R.Kalman, “A new approach to linear filtering and prediction problems”,  Trans. ASME J. Basic Eng., Ser. D,1960, vol. 82, pp. 34–45.
[5]        S.C.Kramer,and H.W.Sorenson, “Recursive Bayesian estimation using piece-wise constant approximations” ,1988, Automatica, vol, 24, no.6, pp.789-801.
[6]        N.Gordon,D.almond,and A.F.M.Smith, “Novel approach to nonlinear and non-Gaussian Bayesian state estimation”, Proc. Inst. Elect. Eng., F, 1993,vol. 140, pp. 107–113.
[7]        A.Doucet, N.Gordon,and V.Krishnamurthy,“Particle filters for state estimation of jump Markov linear systems”, IEEE Trans. Signal Processing,2001,vol. 49, pp. 613–624.
[8]        M.S. Haque, S.Choi,and J.Baek, “Auxiliary Particle Filtering-Based Estimation of Remaining Useful Life of IGBT”", IEEE Transactions on Industrial Electronics, 2018 ,vol.65,no.3,pp. 2693 – 2703.
[9]        A.Murangira,C.Musso,and K.Dahia, “A mixture regularized rao-blackwellized particle filter for terrain positioning, IEEE Transactions on Aerospace and Electronic Systems, 2016 , vol. 52 , no. 4,pp.1967-1985.
[10]     A.Murangira,  M.Christian, C.Musso,and K.Dahia,“A mixture  regularized rao-blackwellized particle filter for terrain positioning” IEEE Transactions on Aerospace and Electronic Systems, 2016 , vol. 52 , no. 4,pp.1967-1985.
[11]     G.Vezzani,U.Pattacini ,G.Battistelli,L.Chisci,and,L.Natale,“ Memory Unscented Particle Filter for 6-DOF Tactile Localization”, IEEE Transactions on Robotics ,2017 , vol. 33 , no. 5, pp.1139 – 1155.
[12]     Y.Zhang,S.Wang,and J.Li, “Improved particle filtering techniques based on generalized interactive genetic algorithm”, Journal of Systems Engineering and Electronics,2016 , vol. 27 , no. 1, pp.242 - 250
[13]     K.Uosaki, Y.Kimura ,and T.Hatanaka, “Nonlinear State Estimation by Evolution Strategies Based Particle Filters”, in: Proc. of the IEEE Congress on Evolutionary Computation,2003.
[14]     G.Tong, Z.Fang ,and X.Xu, “A Particle Swarm Optimized Particle Filter for Nonlinear System State Estimation”, IEEE Congress on Evolutionary Computation Sheraton Vancouver Wall Centre Hotel, Vancouver, BC, Canada ,July 16-21, 2006.
[15]     S.Park, J.Hwang, E. Kim ,and H. Kang , “A New Evolutionary Particle Filter for the Prevention of Sample Impoverishment”, IEEE Trans. On Evolutionary Computation, 2009,vol. 13, no. 4.
[16]      T.Li,M. Bolic, and P.,M. Djuric, “Resampling Methods for Particle Filtering:Classification,implementation,and strategies” , IEEE Signal Processing Magazine, 2015,vol. 32,no.3,pp. 70 – 86.
[17]     G.Choe, T. Wang, F.Liu,S. Hyon, J. Ha, “Particle filter with spline resampling and global transition model”, IET Computer Vision,2015, vol.l.9, no.2, pp.184-197.
[18]     F.Ruknudeen,and S.Asokan, “Application Particle Filter to On-Board Life Estimation of LED Lights”, IEEE Photonics Journal, 2017, vol.9, no.3.
[19]     J. Liu, Z.Wang,and M.Xu, “A Kalman Estimation Based Rao-Blackwellized Particle Filtering for Radar Tracking”, IEEE Access, 2017, vol.5,pp.8162-8174.
[20]     T.Li ,,T.P. Sattar, S.Sun , “Deterministic resampling: Unbiased sampling to avoid sample impoverishment in particle filters”, Signal Processing,20121,vol. 92, no.7, pp.1637–1645.
[21]     A.P.Engelbrecht, “Computational intelligence: An introduction, Second Edition”, Johan Wiley & Sons, Ltd 2007.
[22]     H.Huang, F. Liu, X. Zhuo, and Z. Hao, “ Differential  Evolution Based on Self-Adaptive Fitness Function for Automated Test Case Generation”, IEEE Computational Intelligence Magazine, 2017,  vol.12 ,no.2, pp.46 – 55
[23]     Y.Yu , “Combining H∞ filter and cost-reference particle filter for conditionally linear dynamic systems in unknown non-Gaussian noises”, Signal Processing, 2013,vol.93,no.7, pp.1871–1878.
Journal of Soft Computing and Information Technology
Volume 7, Issue 2 - Serial Number 2
Fall & Winter
January 2019
Pages 16-28

Files

Share

How to cite

Statistics