Particle Filter and Its Application in the Integrated Train Speed Measurement

doi:10.1007/s12204-019-2046-x

Journal of Shanghai Jiao Tong University (Science) ›› 2019, Vol. 24 ›› Issue (1): 130-136.doi: 10.1007/s12204-019-2046-x

• • 上一篇

Particle Filter and Its Application in the Integrated Train Speed Measurement

ZHANG Liang (张梁), BAO Qilian *(鲍其莲), CUI Ke (崔科), JIANG Yaodong (蒋耀东), XU Haigui (徐海贵), DU Yuding (杜雨丁)

(1. Department of Instrument Science and Engineering, School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China; 2. Shanghai Engineering Research Center for Intelligent Diagnosis and Treatment Instrument, Shanghai 200240, China; 3. Xichang Satellite Launch Center, Xichang 615000, Sichuan, China; 4. Department of Research and Development, CASCO Signal Ltd., Shanghai 200240, China)

出版日期:2019-02-28 发布日期:2019-01-28
通讯作者: BAO Qilian *(鲍其莲) E-mail:qlbao@sjtu.edu.cn

Particle Filter and Its Application in the Integrated Train Speed Measurement

ZHANG Liang (张梁), BAO Qilian *(鲍其莲), CUI Ke (崔科), JIANG Yaodong (蒋耀东), XU Haigui (徐海贵), DU Yuding (杜雨丁)

(1. Department of Instrument Science and Engineering, School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China; 2. Shanghai Engineering Research Center for Intelligent Diagnosis and Treatment Instrument, Shanghai 200240, China; 3. Xichang Satellite Launch Center, Xichang 615000, Sichuan, China; 4. Department of Research and Development, CASCO Signal Ltd., Shanghai 200240, China)

Online:2019-02-28 Published:2019-01-28
Contact: BAO Qilian *(鲍其莲) E-mail:qlbao@sjtu.edu.cn

摘要/Abstract

摘要： Particle filter (PF) can solve the problem of state estimation under strong non-linear non-Gaussian noise condition with respect to traditional Kalman filter (KF) and those improved KFs such as extended KF (EKF) and unscented KF (UKF). However, problems such as particle depletion and particle degradation affect the performance of PF. Optimizing the particle set to high likelihood region with intelligent optimization algorithm results in a more reasonable distribution of the sampling particles and more accurate state estimation. In this paper, a novel bird swarm algorithm based PF (BSAPF) is presented. Firstly, different behavior models are established by emulating the predation, flight, vigilance and follower behavior of the birds. Then, the observation information is introduced into the optimization process of the proposal distribution with the design of fitness function. In order to prevent particles from getting premature (being stuck into local optimum) and increase the diversity of particles, L′evy flight is designed to increase the randomness of particle’s movement. Finally, the proposed algorithm is applied to estimate the speed of the train under the condition that the measurement noise of the wheel sensor is non-Gaussian distribution. Simulation study and experimental results both show that BSAPF is more accurate and has more effective particle number as compared with PF and UKF, demonstrating the promising performance of the method.

关键词: particle filter (PF), bird swarm algorithm, fitness function, L′evy flight, proposal distribution, integrated train speed measurement

Abstract: Particle filter (PF) can solve the problem of state estimation under strong non-linear non-Gaussian noise condition with respect to traditional Kalman filter (KF) and those improved KFs such as extended KF (EKF) and unscented KF (UKF). However, problems such as particle depletion and particle degradation affect the performance of PF. Optimizing the particle set to high likelihood region with intelligent optimization algorithm results in a more reasonable distribution of the sampling particles and more accurate state estimation. In this paper, a novel bird swarm algorithm based PF (BSAPF) is presented. Firstly, different behavior models are established by emulating the predation, flight, vigilance and follower behavior of the birds. Then, the observation information is introduced into the optimization process of the proposal distribution with the design of fitness function. In order to prevent particles from getting premature (being stuck into local optimum) and increase the diversity of particles, L′evy flight is designed to increase the randomness of particle’s movement. Finally, the proposed algorithm is applied to estimate the speed of the train under the condition that the measurement noise of the wheel sensor is non-Gaussian distribution. Simulation study and experimental results both show that BSAPF is more accurate and has more effective particle number as compared with PF and UKF, demonstrating the promising performance of the method.

Key words: particle filter (PF), bird swarm algorithm, fitness function, L′evy flight, proposal distribution, integrated train speed measurement

中图分类号:

TP 273

ZHANG Liang (张梁), BAO Qilian *(鲍其莲), CUI Ke (崔科), JIANG Yaodong (蒋耀东), XU Haigu. Particle Filter and Its Application in the Integrated Train Speed Measurement[J]. Journal of Shanghai Jiao Tong University (Science), 2019, 24(1): 130-136.

ZHANG Liang (张梁), BAO Qilian *(鲍其莲), CUI Ke (崔科), JIANG Yaodong (蒋耀东), XU Haigui (徐海贵), DU Yuding (杜雨丁). Particle Filter and Its Application in the Integrated Train Speed Measurement[J]. Journal of Shanghai Jiao Tong University (Science), 2019, 24(1): 130-136.

参考文献 14

[1]	XU S F, XIE L, LIU J L. Robot localization based onMCMC particle filter [J]. Journal of Zhejiang University(Engineering Science), 2007, 41(7): 1083-1087 (inChinese).
[2]	HAMMERSLEY J M, MORTON K W. Poor man’sMonte Carlo [J]. Journal of the Royal Statistical Society,1954, 16(1): 23-28.
[3]	GORDON N J, SALMOND D J. Novel approach tononlinear and non-Gaussian Bayesian state estimation[J]. IEE Proceedings-F, 1993, 140(2): 107-113.
[4]	HU S Q, JING Z L. Overview of particle filter algorithm[J]. Control and Decision, 2005, 20(4): 361-371(in Chinese).
[5]	ARULAMPALAM M S, MASKELL S, GORDONN J, et al. A tutorial on particle filters for onlinenonlinear/non-Gaussian Bayesian tracking [J]. IEEETransactions on Signal Processing, 2002, 50(2): 174-188.
[6]	MERWE R V D, DOUCET A, FREITAS N D, et al.The unscented particle filter [C]//International Conferenceon Neural Information Processing Systems.Cambridge, USA: MIT Press, 2000: 563-569.
[7]	LI L Q, JI H B, LUO J H. The iterated extendedKalman particle filter [C]//IEEE International Symposiumon Communications and Information Technology.[s.l.]: IEEE, 2005: 1213-1216.
[8]	CHEN Z M, TIAN M C, WU P L, et al. Intelligentparticle filter based on bat algorithm [J]. Acta PhysicaSinica, 2017, 66(5): 41-50 (in Chinese).
[9]	MENG X B, GAO X Z, LU L H, et al. A new bioinspiredoptimisation algorithm: Bird swarm algorithm[J]. Journal of Experimental & Theoretical ArtificialIntelligence, 2016, 28(4): 673-687.
[10]	ZHANG Q T, FANG L Q, ZHAO Y L. Double subgroupsfruit fly optimization algorithm with characteristicsof L′evy flight [J]. Journal of Computer Applications,2015, 33(5): 1348-1352 (in Chinese).
[11]	WU B C. Research and application of particle filterresampling algorithms [D]. Harbin, China: School ofComputer Science and Technology, Harbin Institute ofTechnology, 2006 (in Chinese).
[12]	MANTEGNA R N. Fast, accurate algorithm for numericalsimulation of L′evy stable stochastic process[J]. Physical Review E, 1994, 49(5): 4677-4683.
[13]	SPIRYAGIN M, POLACH O, COLE C. Creep forcemodelling for rail traction vehicles based on the Fastsimalgorithm [J]. Vehicle System Dynamics, 2013,55(11): 1765-1783.
[14]	LI Y F. Research on the adhesion control methodsbased on the optimal creep rate [D]. Chengdu,China: Department of Power Electronics and ElectricalDrives, Southwest Jiao Tong University, 2011 (inChinese).

Particle Filter and Its Application in the Integrated Train Speed Measurement

Particle Filter and Its Application in the Integrated Train Speed Measurement

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献 14

相关文章 1

编辑推荐

Metrics

本文评价