Bayesian Multiple Target Track 2nd Ed
This second edition has undergone substantial revision from the 1999 first edition, recognizing that a lot has changed in the multiple target tracking field. One of the most dramatic changes is in the widespread use of particle filters to implement nonlinear, non-Gaussian Bayesian trackers. This book views multiple target tracking as a Bayesian inference problem. Within this framework it develops the theory of single target tracking. In addition to providing a detailed description of a basic particle filter that implements the Bayesian single target recursion, this resource provides numerous examples that involve the use of particle filters.
Why Read This Book
You will gain a modern, Bayesian perspective on the multiple-target tracking problem that emphasizes practical solutions for nonlinear, non-Gaussian systems. The book teaches you how to implement and tune particle-filter-based trackers through clear theory and rich examples, making it ideal if you need robust tracking for radar, sonar, or multisensor systems.
Who Will Benefit
Practicing engineers, graduate students, and researchers in radar/sonar/communications and sensor-fusion who need to design or evaluate Bayesian and particle-filter tracking systems.
Level: Advanced — Prerequisites: Probability and stochastic processes, linear algebra, basic estimation theory (Kalman filter familiarity helpful), and basic programming (MATLAB, Python, or C/C++).
Key Takeaways
- Formulate single- and multiple-target tracking problems within a Bayesian inference framework.
- Implement the Bayesian single-target recursion and practical particle-filter algorithms for nonlinear, non-Gaussian models.
- Design and manage data-association, track initiation, and track termination strategies for multi-target scenarios.
- Evaluate tracker performance and understand computational trade-offs in real sensor systems (radar/sonar/communications).
- Adapt particle filters and importance-sampling strategies to realistic measurement and motion models.
- Apply advanced multi-target methods (e.g., multi-hypothesis concepts and related approximations) and understand when to use them.
Topics Covered
- 1. Introduction: Bayesian View of Tracking
- 2. Probability, Bayesian Inference, and Notation
- 3. Single-Target Bayesian Recursion
- 4. Linear/Gaussian Tracking and the Kalman Filter (review)
- 5. Nonlinear and Non-Gaussian Tracking: Need for Particle Methods
- 6. Particle Filters: Algorithms and Implementation
- 7. Practical Particle Filter Examples and Case Studies
- 8. Multiple-Target Models and Problem Formulation
- 9. Data Association: MHT, JPDA and Bayesian Approaches
- 10. Track Initiation, Maintenance, and Termination
- 11. Performance Evaluation, Computational Issues, and Tuning
- 12. Advanced Topics: PHD/CPHD, Approximate Bayesian Methods
- 13. Practical Considerations, Simulations, and Worked Examples
- Appendices: Mathematical Background and Algorithmic Details
Languages, Platforms & Tools
How It Compares
Complements Bar-Shalom & Li's Multitarget-Multisensor Tracking by shifting focus from classical Gaussian/JPDA/MHT approaches to Bayesian sequential Monte Carlo (particle-filter) solutions; for particle-filter fundamentals see Thrun/Burgard/Fox's Probabilistic Robotics but Stone is tracking-focused.
