General EKF implementation (non-SLAM)
Parent: Extended Kalman Filter
Source: SLAM for Dummies General (non-SLAM) implementation of EKF:
- only state estimation
- robot is given a perfect map
- no map update necessary
SLAM implementations of EKF requires map update and therefore the matrices are changed.
Source: Scaradozzi 2018 SLAM application in surgery EKF vs KF
- circumvents linearity assumption
- uses nonlinear functions to describe
- the next state probability
- measurement probability
- approximates the state distribution with a Gaussian Random Variable