A Bayesian approach to positioning and tracking applications naturally leads to a recursive estimation formulation. The recently invented particle filter provides a numerical solution to the non-tractable recursive Bayesian estimation problem. As an alternative, traditional methods such as the extended Kalman filter. which is based on a linearized model and an assumption on Gaussian noise, yield approximate solutions.

In many practical applications, signal quantization and algorithmic complexity are fundamental issues. For measurement quantization, estimation performance is analyzed in detail. The algorithmic complexity is addressed for the marginalized particle filter, where the Kalman filter solves a linear subsystem subject to Gaussian noise efficiently.

The particle filter is adopted to several positioning and tracking applications and compared to traditional approaches. Particularly, the use of external database information to enhance estimation performance is discussed. In parallel, fundamental limits are derived analytically or numerically using the Cramér-Rao lower bound, and the result from estimation studies is compared to the corresponding lower bound. A framework for map-aided positioning at sea is developed, featuring an underwater positioning system using depth information and readings from a sonar sensor and a novel surface navigation system using radar measurements and sea chart information. Bayesian estimation techniques are also used to improve position accuracy for an industrial robot. The bearings-only tracking problem is addressed using Bayesian techniques and map information is used to improve the estimation performance. For multiple-target tracking problems data association is an important issue. A method to incorporate classical association methods when the estimation is based on the particle filter is presented. A real-time implementation of the particle filter as well as hypothesis testing is introduced for a collision avoidance application.