H-Infinity Filter Based Particle Filter for Maneuvering Target Tracking

Wang, Qicong; Li, Jing; Zhang, Meixiang; Yang, and Chenhui

doi:10.2528/pierb11031504

Cited by 19 publications

(18 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The main idea of particle filter is to represent the probability density of system state by a set of particles with associated weights. It was shown in literatures [6][7][8][9][10] that particle filter is particularly suitable for estimating the state of nonlinear, nonGaussian dynamic system.…”

Section: Introductionmentioning

confidence: 99%

Fuzzy-Control-Based Particle Filter for Maneuvering Target Tracking

Wang¹,

Chen²,

Shi³

et al. 2011

PIER

View full text Add to dashboard Cite

Abstract-In this paper, we propose a novel fuzzy-control-based particle filter (FCPF) for maneuvering target tracking, which combines the advantages of standard particle filter (SPF) and multiple model particle filter (MMPF). That is, the SPF is adopted during non-maneuvering movement while the MMPF is adopted during maneuvering movement. The key point of the FCPF is to use a fuzzy controller, which could imitate the thoughts of human beings in some degree, to detect the target's maneuver and use a backward correction sub-algorithm to alleviate the performance degradation of MMPF caused by detection delay. Simulation results indicate that the proposed algorithm has a much better tracking accuracy than the SPF while keeps approximately equal computational complexity. Compared with MMPF, both algorithms have no tracking lost, but the tracking accuracy of the proposed FCPF is a little better than the MMPF, and the FCPF consumes about 66% computation time of the MMPF. Thus, the proposed algorithm offers a more effective way for maneuvering target tracking.

show abstract

Section: Introductionmentioning

confidence: 99%

Fuzzy-Control-Based Particle Filter for Maneuvering Target Tracking

Wang¹,

Chen²,

Shi³

et al. 2011

PIER

View full text Add to dashboard Cite

show abstract

“…The key idea is to represent the target posterior probability density function (PDF) of the state given the observations by a set of random particles with their associated weights. PFs have shown great promise in various target tracking problems [20][21][22][23][24][25][26][27] since they were proposed. However, for MTT problems, multi-target PF requires intensive computations and it is difficult to find an efficient importance density for the multitarget PF [15].…”

Section: Introductionmentioning

confidence: 99%

Simplified Particle PHD Filter for Multiple-Target Tracking: Algorithm and Architecture

Hong¹,

Wang²,

Shi

et al. 2011

PIER

View full text Add to dashboard Cite

Abstract-In this paper, we propose a simplified particle probability hypothesis density (PHD) filter and its hardware implementation for multiple-target tracking (MTT). In the proposed algorithm, the update step of particle PHD filter is simplified and the time-varying number of measurements is arranged in combination series/parallel mode. This may result in fixed hardware architecture and therefore present a convenient hardware implementation of particle PHD filter. Simulation results indicate that for the MTT problems, this proposed simplified algorithm shows similar performance with the standard particle PHD filter but has faster processing rate. Experiment study shows that the proposed simplified algorithm can be efficiently implemented in hardware and can effectively solve the MTT problems.

show abstract

“…The main problem of MTT is how to deal with the unknown fast change in the maneuvering acceleration. Many techniques and methods [2][3][4][5] have been suggested to solve the problem during last years. Acceleration modeling techniques [6], input estimation (IE) techniques [7] and multiple-model (MM) methods [8,9] are three main approaches.…”

Section: Introductionmentioning

confidence: 99%

Absolute Adaptive Cs Model and Modified Strong Tracking Unscented Filter for High Maneuvering Target Tracking

Zhou¹,

Jin-mang²,

Liu³

2013

PIER C

View full text Add to dashboard Cite

Abstract-Absolute adaptive current statistical (AACS) model and modified strong tracking unscented filter (MSTUF) are proposed for maneuvering target tracking (MTT) under nonlinear measurement in this paper. The key point of the AACS model is to associate the instantaneous acceleration variance with some elements of state covariance matrix by constructing acceleration increment models of the acceleration limit and acceleration mean in the CS model, while the maneuvering frequency can adjust itself according to the change of the measurement residual. MSTUF is proposed for high maneuver tracking under nonlinear measurement by incorporating the modified strong tracking filter (STF) into the unscented filter (UF). Since the state covariance, process noise covariance and maneuvering frequency can adjust themselves jointly according to the residual, the proposed algorithm, called the AACS-MSTUF, has a good performance on both maneuver and non-maneuver. Simulation results indicate that the overall performance of the proposed algorithm is better than the interacting multiple-model unscented filter (IMM-UF), UF and original strong tracking unscented filter (STUF) based on the CS model (CS-STUF) when considering tracking accuracy, stability, convergence and computational complexity.

show abstract

H-Infinity Filter Based Particle Filter for Maneuvering Target Tracking

Cited by 19 publications

References 22 publications

Fuzzy-Control-Based Particle Filter for Maneuvering Target Tracking

Fuzzy-Control-Based Particle Filter for Maneuvering Target Tracking

Simplified Particle PHD Filter for Multiple-Target Tracking: Algorithm and Architecture

Absolute Adaptive Cs Model and Modified Strong Tracking Unscented Filter for High Maneuvering Target Tracking

Contact Info

Product

Resources

About