Skip to main content
State estimation and mapping treat a robot’s belief about its pose and its world as a probability distribution that gets updated as new measurements arrive. The framing follows Thrun, Burgard, and Fox’s Probabilistic Robotics, and it underlies everything from localization in a known map to building a map while moving through it. This chapter covers the recursive Bayesian machinery behind localization and mapping, then walks through the families of filters used in practice, and ends with simultaneous localization and mapping.

State estimation

Recursive Bayesian estimation, Bayes filters, Kalman filters, and HMM-based localization.

Recursive state estimation

The general recursive Bayes filter and its assumptions.

Discrete Bayesian filter

The Bayes filter on a finite state space, with a worked grid-world example.

Kalman filters

Optimal linear-Gaussian estimation and its extensions to nonlinear systems.

HMM localization

Hidden Markov model formulation of robot localization.

Occupancy mapping

Building probabilistic grid maps from range-sensor measurements.

SLAM

Simultaneous localization and mapping: estimating pose and map together.
Edit this page on GitHub or file an issue.