Basic Extended Kalman Filter – Simultaneous Localisation and Mapping

摘要

EKF is a good way to learn about SLAM because of its simplicity whereas probabilistic methods are complex but they handle uncertainty better. This chapter presents some of the basics feature based EKF-SLAM technique used for generating robot pose estimates together with positions of features in the robot's operating environment. It highlights some of the basics for successful EKF ­ SLAM implementation:, these include: Process and observation models, Basic EKF-SLAM Steps, Feature Extraction and Environment modelling, Data Association, and Multi ­ Robot ­ EKF ­ SLAM with more emphasis on the Cooperative