To introduce the major concepts, we will begin with an intuitive description of Markov localization, followed by a mathematical derivation of the algorithm. The reader may notice that Markov localization is a special case of probabilistic state estimation, applied to mobile robot localization (see also [Russell & Norvig1995, Fox1998, Koenig & Simmons1998]).

For clarity of the presentation, we will initially make the
restrictive assumption that the environment is *static*. This
assumption, called *Markov assumption*, is commonly made in the
robotics literature. It postulates that the robot's location is the
only state in the environment which systematically affects sensor
readings. The Markov assumption is violated if robots share the same
environment with people. Further below, in Section 3.3, we
will side-step this assumption and present a Markov localization
algorithm that works well even in highly dynamic environments, e.g.,
museums full of people.

- The Basic Idea
- Basic Notation
- Recursive Localization
- The Markov Localization Algorithm
- Implementations of Markov Localization

Fri Nov 19 14:29:33 MET 1999