State Estimation

State estimation refers to the process of determining the current state of a system based on measurements and a model of the system dynamics.

States can be anything, it doesn’t have to be a position. Ex: In the context of aircrafts, you can do state estimation for the fuel level and engine performance.

Localization is the most popular use case of State Estimation.

Concepts

• Bayes Filter is really good for state estimation
• Also Kalman Filter

Categorizations

Seen from Visual SLAM book:

Whether the motion and observation equations are linear and whether the noise is assumed to be Gaussian, it is divided into

• linear/nonlinear and
• Gaussian/non-Gaussian systems