Integrated Tracking, Classification, and Sensor Management

MAHENDRA MALLICK, PhD, is Principal Research Scientist at the Propagation Research Associates, Inc. A senior member of the IEEE, he has served as the associate editor-in-chief of the online journal of the International Society of Information Fusion (ISIF).VIKRAM KRISHNAMURTHY, PhD, holds the Canada Research Chair in Statistical Signal Processing at The University of British Columbia. He is an IEEE Fellow and Editor-in-Chief of the IEEE Journal of Selected Topics in Signal Processing.BA-NGU VO, PhD, is Professor and Chair of Signals and Systems in the Department of Electrical and Computer Engineering at Curtin University in Western Australia. He is Associate Editor for IEEE Transactions on Aerospace and Electronic Systems.

• PREFACE xviiCONTRIBUTORS xxiiiPART I FILTERING1. Angle-Only Filtering in Three Dimensions 3Mahendra Mallick, Mark Morelande, Lyudmila Mihaylova, Sanjeev Arulampalam, and Yanjun Yan1.1 Introduction 31.2 Statement of Problem 61.3 Tracker and Sensor Coordinate Frames 61.4 Coordinate Systems for Target and Ownship States 71.5 Dynamic Models 91.6 Measurement Models 141.7 Filter Initialization 151.8 Extended Kalman Filters 171.9 Unscented Kalman Filters 191.10 Particle Filters 231.11 Numerical Simulations and Results 281.12 Conclusions 312. Particle Filtering Combined with Interval Methods for Tracking Applications 43Amadou Gning, Lyudmila Mihaylova, Fahed Abdallah, and Branko Ristic 2.1 Introduction 432.2 Related Works 442.3 Interval Analysis 462.4 Bayesian Filtering 512.5 Box Particle Filtering 522.6 Box Particle Filtering Derived from the Bayesian Inference Using a Mixture of Uniform Probability Density Functions 562.7 Box-PF Illustration over a Target Tracking Example 652.8 Application for a Vehicle Dynamic Localization Problem 672.9 Conclusions 713. Bayesian Multiple Target Filtering Using Random Finite Sets 75Ba-Ngu Vo, Ba-Tuong Vo, and Daniel Clark3.1 Introduction 753.2 Overview of the Random Finite Set Approach to Multitarget Filtering 763.3 Random Finite Sets 813.4 Multiple Target Filtering and Estimation 853.5 Multitarget Miss Distances 913.6 The Probability Hypothesis Density (PHD) Filter 953.7 The Cardinalized PHD Filter 1053.8 Numerical Examples 1113.9 MeMBer Filter 1174. The Continuous Time Roots of the Interacting Multiple Model Filter 127Henk A.P. Blom4.1 Introduction 1274.2 Hidden Markov Model Filter 1294.3 System with Markovian Coefficients 1364.4 Markov Jump Linear System 1414.5 Continuous-Discrete Filtering 1494.6 Concluding Remarks 154PART II MULTITARGET MULTISENSOR TRACKING5. Multitarget Tracking Using Multiple Hypothesis Tracking 165Mahendra Mallick, Stefano Coraluppi, and Craig Carthel5.1 Introduction 1655.2 Tracking Algorithms 1665.3 Track Filtering 1705.4 MHT Algorithms 1795.5 Hybrid-State Derivations of MHT Equations 1805.6 The Target-Death Problem 1855.7 Examples for MHT 1865.8 Summary 1896. Tracking and Data Fusion for Ground Surveillance 203Michael Mertens, Michael Feldmann, Martin Ulmke, and Wolfgang Koch6.1 Introduction to Ground Surveillance 2036.2 GMTI Sensor Model 2046.3 Bayesian Approach to Ground Moving Target Tracking 2096.4 Exploitation of Road Network Data 2226.5 Convoy Track Maintenance Using Random Matrices 2346.6 Convoy Tracking with the Cardinalized Probability Hypothesis Density Filter 2437. Performance Bounds for Target Tracking: Computationally Efficient Formulations and Associated Applications 255Marcel Hernandez7.1 Introduction 2557.2 Bayesian Performance Bounds 2587.3 PCRLB Formulations in Cluttered Environments 2627.4 An Approximate PCRLB for Maneuevring Target Tracking 2697.5 A General Framework for the Deployment of Stationary Sensors 2717.6 UAV Trajectory Planning 2947.7 Summary and Conclusions 3058. Track-Before-Detect Techniques 311Samuel J. Davey, Mark G. Rutten, and Neil J. Gordon8.1 Introduction 3118.2 Models 3188.3 Baum Welch Algorithm 3278.4 Dynamic Programming: Viterbi Algorithm 3318.5 Particle Filter 3348.6 ML-PDA 3378.7 H-PMHT 3418.8 Performance Analysis 3478.9 Applications: Radar and IRST Fusion 3548.10 Future Directions 3579. Advances in Data Fusion Architectures 363Stefano Coraluppi and Craig Carthel9.1 Introduction 3639.2 Dense-Target Scenarios 3649.3 Multiscale Sensor Scenarios 3689.4 Tracking in Large Sensor Networks 3709.5 Multiscale Objects 3729.6 Measurement Aggregation 3789.7 Conclusions 38310. Intent Inference and Detection of Anomalous Trajectories: A Metalevel Tracking Approach 387Vikram Krishnamurthy10.1 Introduction 38710.2 Anomalous Trajectory Classification Framework 39310.3 Trajectory Modeling and Inference Using Stochastic Context-Free Grammars 39510.4 Trajectory Modeling and Inference Using Reciprocal Processes (RP) 40310.5 Example 1: Metalevel Tracking for GMTI Radar 40610.6 Example 2: Data Fusion in a Multicamera Network 40710.7 Conclusion 413PART III SENSOR MANAGEMENT AND CONTROL11. Radar Resource Management for Target Tracking—A Stochastic Control Approach 417Vikram Krishnamurthy11.1 Introduction 41711.2 Problem Formulation 42211.3 Structural Results and Lattice Programming for Micromanagement 43111.4 Radar Scheduling for Maneuvering Targets Modeled as Jump Markov Linear System 43711.5 Summary 44412. Sensor Management for Large-Scale Multisensor-Multitarget Tracking 447Ratnasingham Tharmarasa and Thia Kirubarajan12.1 Introduction 44712.2 Target Tracking Architectures 45112.3 Posterior Cram´er–Rao Lower Bound 45212.4 Sensor Array Management for Centralized Tracking 45812.5 Sensor Array Management for Distributed Tracking 47312.6 Sensor Array Management for Decentralized Tracking 48912.7 Conclusions 507PART IV ESTIMATION AND CLASSIFICATION13. Efficient Inference in General Hybrid Bayesian Networks for Classification 523Wei Sun and Kuo-Chu Chang13.1 Introduction 52313.2 Message Passing: Representation and Propagation 52613.3 Network Partition and Message Integration for Hybrid Model 53213.4 Hybrid Message Passing Algorithm for Classification 53613.5 Numerical Experiments 53713.6 Concluding Remarks 54414. Evaluating Multisensor Classification Performance with Bayesian Networks 547Eswar Sivaraman and Kuo-Chu Chang14.1 Introduction 54714.2 Single-Sensor Model 54814.3 Multisensor Fusion Systems—Design and Performance Evaluation 56014.4 Summary and Continuing Questions 56415. Detection and Estimation of Radiological Sources 579Mark Morelande and Branko Ristic15.1 Introduction 57915.2 Estimation of Point Sources 58015.3 Estimation of Distributed Sources 59015.4 Searching for Point Sources 59915.5 Conclusions 612PART V DECISION FUSION AND DECISION SUPPORT16. Distributed Detection and Decision Fusion with Applications to Wireless Sensor Networks 619Qi Cheng, Ruixin Niu, Ashok Sundaresan, and Pramod K. Varshney16.1 Introduction 61916.2 Elements of Detection Theory 62016.3 Distributed Detection with Multiple Sensors 62416.4 Distributed Detection in Wireless Sensor Networks 63416.5 Copula-Based Fusion of Correlated Decisions 64516.6 Conclusion 65217. Evidential Networks for Decision Support in Surveillance Systems 661Alessio Benavoli and Branko Ristic17.1 Introduction 66117.2 Valuation Algebras 66217.3 Local Computation in a VA 66817.4 Theory of Evidence as a Valuation Algebra 67217.5 Examples of Decision Support Systems 685References 702Index 705