Performance evaluation of multitarget tracking algorithms

When tracking multiple targets using multiple sensors, the performance evaluation of different estimation and data association algorithms becomes critical in system design and global level coordination. One of the most important problems in tracking multiple targets is the measurement origin uncertainty, i.e., one has to deal with missed detections and false alarms when trying to assign uncertain origin measurements to tracks. To better understand the theoretical bound of the parameter estimation of a single target in the presence of clutter, the maximum likelihood estimator with probabilistic data association (ML-PDA) is formulated and the Cramer Rao lower bound (CRLB) is obtained. It is shown that a scalar information reduction factor (IRF) quantifies the information loss due to measurement origin uncertainty. To generalize the ML-PDA technique to the parameter estimation of multiple targets, one needs to determine the number of targets in the observation space based on a certain model selection criterion. A new approach, using the minimum description length (MDL) criterion, is presented for the detection and initiation of multiple tracks with the ML-PDA estimator. The tracking performance using multiple sensors is studied on two data processing algorithms, namely, the centralized filter and a distributed scheme relying on track-to-track fusion. Based on a simple target kinematic model, it is shown that the optimal Huimin Chen––University of Connecticut, 2002 track-to-track fusion is indeed suboptimal compared with the centralized solution and the performance difference between the centralized and distributed trackers increases as the number of sensors increases even under the assumption of perfect data association. Finally, data association with covariance modification is studied under the top-M multiple hypothesis tracking (TMHT) framework. The N dimensional assignment problem is formulated including possibly unresolved measurements due to the finite resolution of a sensor when tracking closely-spaced targets. The M-best solutions of the N dimensional assignment are used to modify the covariance of each track based on the probabilistic data association filter (PDAF) at the rear end of the sliding window. A suboptimal approach using a linear program to solve the relaxed assignment problem is also compared with single best assignment solution and TMHT. 2002 ii ACKNOWLEDGEMENTS It is my pleasure to have this opportunity to thank those, who, in different ways, have contributed to this dissertation. My major advisor, Professor Yaakov Bar-Shalom, deserves my deep appreciation for his excellent guidance. His scholarship in conducting research, his insight and expertise in the tracking field, as well as …

[1]  G. Schwarz.  Estimating the Dimension of a Model , 1978 .

[2]  Jorma Rissanen,et al.  Hypothesis Selection and Testing by the MDL Principle , 1999, Comput. J..

[3]  Jorma Rissanen,et al.  Strong optimality of the normalized ML models as universal codes and information in data , 2001, IEEE Trans. Inf. Theory.

[4]  l.V. Bojilov,et al.  DATA ASSOCIATION ALGORITHM FOR MULTISENSOR MULTITARGET TRACKING , 2004 .

[5]  Yaakov Bar-Shalom,et al.  Ballistic missile track initiation from satellite observations , 1994, Defense, Security, and Sensing.

[6]  Thiagalingam Kirubarajan,et al.  Performance limits of track-to-track fusion versus centralized estimation: theory and application [sensor fusion] , 2003 .

[7]  D. D. Mueller,et al.  Fundamentals of Astrodynamics , 1971 .

[8]  Thiagalingam Kirubarajan,et al.  Estimation with Applications to Tracking and Navigation , 2001 .

[9]  Ted Chang.  Geometrical foundations of asymptotic inference , 2002 .

[10]  Tian Zhi,et al.  Performance Evaluation of Track Fusion with Information , 2002 .

[11]  Krishna R. Pattipati,et al.  IMM estimation for multitarget-multisensor air traffic surveillance , 1997 .

[12]  Y. Bar-Shalom,et al.  A new relaxation algorithm and passive sensor data association , 1992 .

[13]  G. A. Watson,et al.  IMMPDAF for radar management and tracking benchmark with ECM , 1998 .

[14]  Y. Bar-Shalom,et al.  Track formation with bearing and frequency measurements in clutter , 1990, 29th IEEE Conference on Decision and Control.

[15]  S. Kay.  Fundamentals of statistical signal processing: estimation theory , 1993 .

[16]  G. A. Watson,et al.  Benchmark Problem with a Multisensor Framework for Radar Resource Allocation and the Tracking of Highly Maneuvering Targets, Closely Spaced Targets, and Targets in the Presence of Sea-Surface4nduced Multipath (CD-ROM) , 1999 .

[17]  Huimin Chen,et al.  One-step solution for the general out-of-sequence-measurement problem in tracking , 2002, Proceedings, IEEE Aerospace Conference.

[18]  I. J. Myung,et al.  Counting probability distributions: Differential geometry and model selection , 2000, Proc. Natl. Acad. Sci. USA.

[19]  W. Blair,et al.  Unresolved Rayleigh target detection using monopulse measurements , 1998 .

[20]  Peter Willett,et al.  Matrix CRLB scaling due to measurements of uncertain origin , 2001, IEEE Trans. Signal Process..

[21]  Y. Bar-Shalom,et al.  On optimal track-to-track fusion , 1997, IEEE Transactions on Aerospace and Electronic Systems.

[22]  J. A. Roecker,et al.  Comparison of two-sensor tracking methods based on state vector fusion and measurement fusion , 1988 .

[23]  Kuo-Chu Chang,et al.  Joint probabilistic data association for multitarget tracking with possibly unresolved measurements and maneuvers , 1984 .

[24]  X. R. Li,et al.  Optimal Linear Estimation Fusion — Part III : Cross-Correlation of Local Estimation Errors , 2001 .

[25]  Yunmin Zhu,et al.  Unified optimal linear estimation fusion. I. Unified models and fusion rules , 2000, Proceedings of the Third International Conference on Information Fusion.

[26]  R. Kass,et al.  Geometrical Foundations of Asymptotic Inference , 1997 .

[27]  Jorma Rissanen,et al.  Fisher information and stochastic complexity , 1996, IEEE Trans. Inf. Theory.

[28]  Krishna R. Pattipati,et al.  MDL approach for multiple low-observable track initiation , 2002, SPIE Defense + Commercial Sensing.

[29]  Steven Kay,et al.  Fundamentals Of Statistical Signal Processing , 2001 .

[30]  Yaakov Bar-Shalom,et al.  Benchmark for radar allocation and tracking in ECM , 1998 .

[31]  W. Koch,et al.  Multiple hypothesis track maintenance with possibly unresolved measurements , 1997, IEEE Transactions on Aerospace and Electronic Systems.

[32]  J. Rissanen.  Stochastic complexity and the mdl principle , 1987 .

[33]  Dimitri P. Bertsekas,et al.  Network optimization : continuous and discrete models , 1998 .

[34]  Thiagalingam Kirubarajan,et al.  Precision large scale air traffic surveillance using IMM/assignment estimators , 1999 .

[35]  Oliver E. Drummond,et al.  Track fusion with feedback , 1996, Defense, Security, and Sensing.

[36]  Thiagalingam Kirubarajan,et al.  Maximum likelihood angle extractor for two closely spaced targets , 2002 .

[37]  Oliver E. Drummond,et al.  Multiple target tracking with multiple frame, probabilistic data association , 1993, Defense, Security, and Sensing.

[39]  Y. Bar-Shalom,et al.  Low observable target motion analysis using amplitude information , 1995, Proceedings of 1995 American Control Conference - ACC'95.

[40]  Zhi-Quan Luo,et al.  An interior point linear programming approach to two-scan data association , 1999 .

[41]  X. R. Li,et al.  Optimal Linear Estimation Fusion — Part IV : Optimality and Efficiency of Distributed Fusion , 2001 .

[42]  Vijay Balasubramanian,et al.  A Geometric Formulation of Occam's Razor For Inference of Parametric Distributions , 1996, adap-org/9601001.

[43]  H. Akaike.  A new look at the statistical model identification , 1974 .

[44]  X. R. Li,et al.  Unified Optimal Linear Estimation Fusion — Part I : Unified Models and Fusion Rules , 2000 .

[45]  S. Sivananthan,et al.  A radar power multiplier algorithm for acquisition of low observable ballistic missiles using an ESA radar , 1999, 1999 IEEE Aerospace Conference. Proceedings (Cat. No.99TH8403).

[46]  H. Bozdogan.  On the information-based measure of covariance complexity and its application to the evaluation of multivariate linear models , 1990 .

[47]  Yaakov Bar-Shalom,et al.  Detection and estimation for multiple targets with two omnidirectional sensors in the presence of false measurements , 1990, IEEE Trans. Acoust. Speech Signal Process..