This is a reprint of the book originally published by Artech House in 1993, following the transfer of copyright to YBS Publishing. This comprehensive, practical source provides the engineering tools for the design and evaluation of state estimators in stochastic systems. It is a complete guide to the technology fundamentals that will be used in tomorrow's civil and military systems and saves you valuable time because it provides ready access to the analytic tools that allow you to design and evaluate state estimators. It is the first text to profile guidelines for the selection of estimator design parameters for standard estimators, such as the Kalman filter, as well as the latest multiple estimators, including the recently developed Interacting Multiple Model algorithm, an ideal self-adjusting variable bandwidth estimator for hybrid systems --- with multiple modes of behavior. This book doesn't waste time discussing only the theory behind estimation and tracking. It shows you realistic applications --- such as air traffic tracking --- that will add new dimensions to your design thinking.
Included with Estimation and Tracking is the MATLAB software DynaEst 2.1, an interactive design tool for Kalman filters and an adaptive multiple model estimator. The user may create models for all dimensions and may specify any parameter of the true models and/or the filter assumed models, such as noise means and variances. Most relevant statistical tests of the Kalman filters may be performed, including the normalized estimation error squared (NEES), the normalized innovation squared (NIS), and the RMS errors. DynaEst provides valuable hands-on experience with Kalman filters, and some practical problems of estimation and tracking can be handled by this software. DynaEst includes the following convenient, user-friendly features: Self-explanatory menu; Convenient user specification of parameters; Plotting/printing on screen, to a printer, or to a file; Automatic installation; Access to DOS and MATLAB from within DynaEst; Easily readable source code transparent to the user (user modifiable); Running on a floppy diskette as well as on a hard disk; Knowledge of MATLAB not required. With 1,700 equations and 50 figures as well as the software support, it is a practical source of design techniques for engi- neers/scientists as well as graduate students in aerospace systems, control systems, and robotics.
Professors will appreciate the ease and versatility of the book's large-sized textgraph format which allows the class lecturer to reproduce the text's pages as transparencies for overhead viewgraphs. It is also a must-have for applied researchers and design engineers in civil and military surveillance systems.
Here are some reviews.
IBM compatible PC DOS 5.0 or higher. MATLAB (professional, education, or student version).
Introduction. Brief Review of Linear Algebra and Linear Systems.
Brief Review of Probability Theory.
Brief Review of Statistics.
Some Basic Concepts in Estimation.
Linear Estimation in Static Systems.
Linear Dynamic Systems with Random Inputs.
State Estimation in Discrete-Time Linear Dynamic Systems.
Estimation for Kinematic Models.
Computational Aspects of Estimation.
Extensions of Discrete-Time Estimation.
Continuous-Time Linear State Estimation.
State Estimation for Nonlinear Dynamic Systems.
Adaptive Estimation and Maneuvering Targets.
PROBLEM SOLUTIONS: AVAILABLE TO INSTRUCTORS UPON REQUEST.
1998. ISBN: 096483121X. Ring Binder. 8 1/2'' x 11''. 525 pages. 1 diskette. Cost: $65.00 .
How to order from the publisher: Book rate shipping within US is free. Shipping outside US is $5/book by surface mail and $15/book by airmail. To order, send the form below with a check (drawn on a US bank) to
Storrs, CT 06269-2157
How to order this book from the distributor: Credit card charges can be placed at amazon.com.
Book: E&T Number of copies: Cost (include shipping as appropriate):
Book: MTMST-I Number of copies: Cost (include shipping as appropriate):
Book: MTMST-II Number of copies: Cost (include shipping as appropriate):
(mtt/et&aapub June 4, 1998)
This is a reprint of the book originally published by Artech House in 1990, following the transfer of copyright to YBS Publishing.
"This edited book consists of chapters written by the leading experts in the field of tracking. A no-nonsense book packed with relevant information for the engineer who needs it, an indispensable source of facts and figures for ECM and ECCM engineers." [Microwave Journal]
Here are some reviews.
Development of Practical PDA Logic for Multitarget Tracking by Microprocessor.
Automatic Track Formation in Clutter with a Recursive Algorithm.
Issues in the Design of Practical Multitarget Tracking Algorithms.
Dynamic Programming Algorithm for Detecting Dim Moving Targets.
Registration: A Prerequisite for Multiple Sensor Tracking.
Association and Fusion of Multiple Sensor Data.
Passive Multisensor Data Association Using a New Relaxation Algorithm.
Distributed Multitarget Multisensor Tracking.
Multitarget Tracking Applied to Automatic Target Recognition with an Imaging Infrared Sensor.
Electromechanical Systems for Optical Target Tracking Sensors.
1998. ISBN: 0964831228. Ring Binder. 8 1/2'' x 11''. 390 pages. Cost: $60.00
This is a reprint of the book originally published by Artech House in 1992, following the transfer of copyright to YBS Publishing.
This edited book, second in a series, consists of chapters written by the leading experts in the field of tracking.
Design of a Tracking Algorithm for an Advanced ATC System.
Design of a Multisensor Tracking System for Advanced Air Traffic Control.
Passive Sensor Data Fusion and Maneuvering Target Tracking.
Tracking of Splitting Targets in Clutter by Using an Interacting Multiple Model Joint Probabilistic Data Association Filter.
Precision Tracking of Small Extended Targets with Imaging Sensors.
A System Approach to Multiple Target Tracking.
Performance Analysis of Optimal Data Association with Applications to Multiple-Target Tracking.
Multitarget Tracking with an Agile Beam Radar.
Autonomous Navigation with Uncertain Reference Points Using the PDAF.
The Sensor Management Imperative.
Attribute Fusion and Situation Assessment with a Many-Valued Logic Approach.
1998. ISBN: 0964831236. Ring Binder. 8 1/2'' x 11''. 480 pages. Cost: $60.00
This text (1995) is the most comprehensive compilation of practical algorithms for the estimation of the states of targets in surveillance systems operating in a multitarget-multisensor environment. This problem is characterized by measurement origin uncertainty, typical for low observables. A listing of the various tracking problems and the appropriate algorithms is given and the assumptions and applicability of each algorithm are clearly indicated. The tools for design of algorithms for the association of measurements and tracking are presented. Explicit consideration is given for measurements obtained from different sensors under realistic assumptions --- lack of synchronicity and different detection and accuracy characteristics. Several real-data examples are given to illustrate the techniques discussed. The modeling accounts for target maneuvers, detection probability, false alarms, interference from other targets and the finite resolution capability of sensors. The problems of track initiation, track maintenance and track-to-track association and fusion in a multisensor situation are considered. The optimization of certain signal processing parameters based on tracking performance is also discussed. Many of these techniques have applications to state estimation when using multiple sensors in control systems, autonomous vehicle navigation and robotics. This text is written in textgraph TM format --- a self-contained concise text also suitable for use as viewgraphs. An extensive index is provided with all the indexed terms highlighted in the text for the convenience of the reader.
INTRODUCTION. Estimation, Decision and Tracking in Advanced Surveillance Systems. TRACKING AND DATA ASSOCIATION. OVERVIEW OF TRACKING PROBLEMS AND ALGORITHMS. Track Formation with Single Sensor. Track Maintenance with Single Sensor with Measurement Origin Uncertainty. Track Formation with Multiple Sensors. Track Maintenance with Multiple Sensors. OVERVIEW OF THE BASIC TRACKING ALGORITHMS. The Kalman Filter. Practical issues in tracking not answered by the KF formulation. Alpha-Beta-Gamma Filter. Selection of filter design parameters. Extended Kalman Filter. Multiple Model Approach. The IMM Estimation Algorithm. CONSISTENT COORDINATE CONVERSION IN TRACKING. Validity Limit of the Standard Conversion. EXAMPLE OF DESIGN OF AN IMM ESTIMATOR. A Generic ATC Tracking Problem. Sensitivity to the Design Parameters. ADAPTIVE BEAM POINTING CONTROL OF A PHASED ARRAY RADAR WITH AN IMM ESTIMATOR. Estimator Design for Variable Sampling Interval. BIBLIOGRAPHICAL NOTES. MULTIPLE SENSORS,CLUTTER AND MULTIPLE TARGETS. TRACKING WITH MULTIPLE SENSORS. Synchronous Case. Asynchronous Case. State Update with Out of Sequence Measurements. VALIDATION OF THE MEASUREMENTS. A CLUTTER MODEL. TRACK FORMATION IN CLUTTER. Markov Chain Model. A Design Example. NOTES AND PROBLEMS. ALGORITHMS FOR TRACKING A SINGLE TARGET IN CLUTTER. USE OF THE STANDARD FILTER. Nearest Neighbor and Strongest Neighbor. THE TRACK SPLIT FILTER. THE PROBABILISTIC DATA ASSOCIATION FILTER. Assumptions. A Nonlinear Multisensor Tracking in Clutter Example. Augmented PDA with Feature Measurements. Computational requirements. THE OPTIMAL BAYESIAN APPROACH. Computational Load. TRACKING OF A TARGET IN THE PRESENCE OF WAKE. TRACK FORMATION WITH BEARING AND FREQUENCY MEASUREMENTS IN THE PRESENCE OF FALSE DETECTIONS. The CRLB in the Presence of False Measurements. Low SNR Track Formation with a Passive Sonar. NOTES AND PROBLEMS. MANEUVERING TARGET IN CLUTTER. USE OF THE PDAF WITH A FEATURE MEASUREMENT FOR MANEUVERING TARGET TRACKING. Examples. THE MULTIPLE MODEL PDAF. Application to a Real Target. Adaptive Grid Approach. MUL- TIPLE MODEL TRACK FORMATION AND MAINTENANCE. Track Formation with the IMMPDAF. Estimator Design Parameters. A Real-Data Electro-Optical Sensor Problem. MULTISENSOR TRACKING OF A MANEUVERING TARGET IN CLUTTER. An Example with Sea Skimmers. INTERACTING MULTIPLE MODEL TRACKING WITH TARGET AMPLITUDE FEATURE. The IMMPDAFAI for Track Formation and Maintenance. Coordinated Turn Models. Mixing Models of Different Dimension. Effect of Detection Threshold. System Level Performance Measures. Simulated and Real-Data Examples. NOTES AND PROBLEMS. TRACKING PERFORMANCE PREDICTION AND DETECTION THRESHOLD OPTIMIZATION. TRACKING ACCURACY IN CLUTTER. THE TRACKER OPERATING CHARACTERISTIC. A HYBRID ALGORITHM FOR PREDICTION OF TRACK- ING PERFORMANCE IN CLUTTER. Track Loss Prediction. Detection Threshold Optimization. BIBLIOGRAPHICAL NOTES. MULTIPLE TARGETS IN CLUTTER: BAYESIAN APPROACHES. THE JOINT PROBABILISTIC DATA ASSOCIATION FILTER. The Parametric and Nonparametric JPDA. Passive Sonar Tracking of Two Crossing Targets in Clutter. Coupled Filtering. THE MULTIPLE HYPOTHESIS TRACKING. One Cycle of the MHT Algorithm. Clustering and Hypotheses Reduction Techniques. Example of Track Initiation. Example of Crossing Tracks. MULTIPLE TARGET TRACKING WITH POSSIBLY UNRESOLVED MEASUREMENTS. Update with a Merged Measurement. Application to Crossing Targets. BIBLIOGRAPHICAL NOTES. MULTIPLE TARGETS IN CLUTTER: NON-BAYESIAN APPROACHES. The Assignment Problem. A JOINT LIKELIHOOD FUNCTION METHOD FOR TRACK FORMATION VIA ASSIGNMENT. Partitioning of the Measurements. The Joint Likelihood Ratio. Extension to Incomplete Tracks. Track Formation for Ballistic Objects from a Single Passive Sensor. A LOGIC-BASED MULTITARGET TRACK INITIATOR. Track Start-Up. Track Continuation. Ambiguity Resolution. An Exoatmospheric Ballistic Missile Defense Application. System Level Performance. EVALUATION OF A COMPOSITE LOGIC FOR TRACK FORMATION: A CASE STUDY. The Common Gate-History Algorithm. Automatic Generation of the Markov Chain. Overall System Load. Multiple Targets and False Alarms. Inference on Minimum Target Separation. System Optimization. BIBLIOGRAPHICAL NOTES. MULTISENSOR TRACKING AND DATA FUSION. Decentralized vs. Distributed System. MULTISENSOR TRACKING CON- FIGURATIONS. Single Sensor Tracking Followed by Track Fusion. Static Intersensor Association Followed by Central Processing. Centralized Tracking. A Hierarchical Hybrid Configuration. TESTING AND FUSION OF INDEPENDENT TRACKS. A Fusion Example. TESTING AND FUSION OF DEPENDENT TRACKS. Evaluation of the Effect of the Cross-Covariance on the Inter- sensor Fusion. An Approximation of the Cross-Covariance. MULTISENSOR FUSION OF FEATURE ESTIMATES. The Fusion Equation. Example. DISTRIBUTED APPROACHES TO MULTISENSOR TRACKING WITH OPTIMAL INFORMATION FU- SION. Optimal Distributed Estimation with a Central Estimator. Distributed Estimation without a Central Estimator. Synchronized Sensors. Asynchronous Sensors. MULTISENSOR AIR TRAFFIC SURVEILLANCE --- A CASE STUDY. Design of the Tracking Filter. Measurement Conversions. State Estimate Conversion. Initialization. The 2-D Assignment Problem. Real Data Examples. A MULTISENSOR-MULTITARGET DATA ASSOCIATION ALGORITHM FOR HETEROGENEOUS SENSORS. The Generalized Likelihood Ratio. Generalized 3-D Matching Problem. NP-hardness. The Relaxation Algorithm. Simulation Results. Suboptimality and Association Accuracy. MEASUREMENT ASSOCIATION FROM DIRECTION FINDERS. The S-D Assignment via Lagrangian Relaxation. Example of Localization of Emitters. NOTES AND PROBLEMS. TRACKING WITH IMAGING SENSORS. MEASUREMENT EXTRACTION FOR AN EXTENDED TARGET. Estimation of the Centroid. The Offset Measurement from Image Correlation. Numerical Example and Validation. PRECISION TRACKING OF THE IMAGE CENTROID FOR A SINGLE TARGET. Filter Model for Autocorrelated Measurement Noise. Hot Spot Target Tracking Simulation Results. Relationship to Real Target Parameters. PRECISION TRACKING OF THE IMAGE CENTROIDS FOR TWO TARGETS WITH OVERLAPPING IMAGES. The Centroid of the Combined Image. The Offset Measurement with Two Targets. Example. Targets with Crossing Images --- Simulation Results. PRECISION TRACKING WITH SEGMENTATION FOR IMAGING SENSORS. The Approach to Tracking the Image of Interest. Centroid of a Random Cluster. Clustering and the Centroid Variance. The Optimal Proximity Distance for Clustering. The Segmentation. Measurement Extraction and Tracking. Precision Tracking of a Real Target. Filter Consistency. NOTES AND PROBLEMS. PROBLEM SOLUTIONS. BIBLIOGRAPHY. INDEX (850 entries).
Softcover. 8 1/2" x 11". Approx. 630 pages, 1001 equations, 250 figures, 200 references. ISBN 0-9648312-0-1.
How to order this book from the publisher: The cost per copy is a (nonlinear) function of the number of copies ordered: $120.00 for 1 copy, $110.00 for 2 copies, $100.00 for 3, $90.00 for 4, $85.00 for 5 and $80.00 for 6 or more. This includes shipping within US. Shipping outside US is $5/book by surface mail and $15/book by airmail. To order, send the form below with a personal or company check (drawn on a US bank), or international money order in US$ to
Storrs, CT 06269-2157
How to order this book from the distributor: Single copy orders and credit card charges can be placed at
OPAMP Tech. Books
1033 N. Sycamore Ave.
Los Angeles, CA 90038
Phone: 800-468-4322, 213-464-4322
The cost per copy is $120.00, domestic shipping $4.95. For shipping options and charges contact OPAMP.
ORDER FORM FOR MULTITARGET-MULTISENSOR TRACKING: PRINCIPLES AND TECHNIQUES
Number of copies: Cost (include shipping as appropriate):
Return to Welch and Bishop's Kalman filter page.
Return to Greg Welch's home page.