Track Extraction With Hidden Reciprocal Chains

This technical note (TN) develops Bayesian track extraction algorithms for targets modeled as hidden reciprocal chains (HRC). HRC are a class of finite-state random process models that generalize the familiar hidden Markov chains (HMC). HRC are able to model the “intention” of a target to proceed from a given origin to a destination, behavior that cannot be properly captured by an HMC. While Bayesian estimation problems for HRC have previously been studied, this TN focuses principally on the problem of track extraction, that is, confirming target existence in a set of observations of uncertain origin. Simulation examples are presented, which show that the additional model information contained in an HRC, measured in terms of Kullback–Leibler divergence, improves detection performance when compared to HMC models.

[1]  Langford B. White,et al.  Modelling and Estimation for Finite State Reciprocal Processes , 2012, IEEE Transactions on Automatic Control.

[2]  Alan S. Willsky,et al.  Algorithms for the incorporation of predictive information in surveillance theory , 1985 .

[3]  B. Jamison.  Reciprocal processes , 1974 .

[4]  A. Krener,et al.  Modeling and estimation of discrete-time Gaussian reciprocal processes , 1990 .

[5]  David J. Hand,et al.  Measuring classifier performance: a coherent alternative to the area under the ROC curve , 2009, Machine Learning.

[6]  James Llinas,et al.  Comparative analysis of alternative ground target tracking techniques , 2000, Proceedings of the Third International Conference on Information Fusion.

[7]  Vikram Krishnamurthy,et al.  Detection of Anomalous Trajectory Patterns in Target Tracking via Stochastic Context-Free Grammars and Reciprocal Process Models , 2013, IEEE Journal of Selected Topics in Signal Processing.

[8]  Langford B. White,et al.  Destination-aware target tracking via syntactic signal processing , 2011, 2011 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[9]  Krishna R. Pattipati,et al.  Ground-target tracking with topography-based variable-structure IMM estimator , 1998, Defense, Security, and Sensing.

[10]  Wolfgang Koch,et al.  On Sequential Track Extraction within the PMHT Framework , 2008, EURASIP J. Adv. Signal Process..

[11]  Langford B. White,et al.  New Normalised Bayesian smoothers for signals modelled by non-causal compositions of reciprocal chains , 2014, 2014 IEEE Workshop on Statistical Signal Processing (SSP).

[12]  J.E. Kulkarni,et al.  State-Space Decoding of Goal-Directed Movements , 2008, IEEE Signal Processing Magazine.

[13]  A. Beghi,et al.  On the relative entropy of discrete-time Markov processes with given end-point densities , 1996, IEEE Trans. Inf. Theory.

[14]  S.S. Blackman,et al.  Multiple hypothesis tracking for multiple target tracking , 2004, IEEE Aerospace and Electronic Systems Magazine.

[15]  T. Georgiou,et al.  Positive contraction mappings for classical and quantum Schrödinger systems , 2014, 1405.6650.

[16]  Langford B. White,et al.  Optimal Smoothing for Finite State Hidden Reciprocal Processes , 2011, IEEE Transactions on Automatic Control.

[17]  Langford B. White,et al.  Maximum Likelihood Sequence Estimation for Hidden Reciprocal Processes , 2013, IEEE Transactions on Automatic Control.

[18]  B. Pannetier,et al.  VS-IMM using road map information for a ground target tracking , 2005, 2005 7th International Conference on Information Fusion.

[19]  G. V. Keuk.  Sequential track extraction , 1998 .

[20]  Vikram Krishnamurthy,et al.  Spatiotemporal trajectory models for metalevel target tracking , 2015, IEEE Aerospace and Electronic Systems Magazine.

[21]  Langford B. White,et al.  Normalized Optimal Smoothers for a Class of Hidden Generalized Reciprocal Processes , 2017, IEEE Transactions on Automatic Control.