Localization

Definition of the SLAM problem:

Given:

  • The robot’s controls $$u_{1:T}={u_1, u_2, u_3, …, u_T}$$

  • Observations: $$z_{1:T}={z_1, z_2, z_3, …, z_T}$$ Wanted:

  • Map of the environment

  • Path of the robot $$x_{0:T}={x_0, x_1, x_2, x_T}$$

Estimating the robot’s location.