Intelligent Vehicles Localization Approaches between Estimation and Information: A Review

Abstract

Localization and Simultaneous Localization and Mapping (SLAM) is a fundamental module in any Autonomous vehicle or Autonomous Mobile Robot. In order for an autonomous system to be able to plan, navigate and perform tasks in its environment, it has to be able to localize itself in this environment. This led to a huge amount of research during the past few years in the field of localization and SLAM algorithms. These algorithms are composed of two main parts, a front-end responsible for extracting important features from the sensors data as well as data association. The second part, namely, the back-end is responsible for the probabilistic estimation step. The back-end of these algorithms is heavily dependent on the estimation theory and almost all the related work utilizes one estimation algorithms or another. Throughout this paper we review the most popular estimation algorithms used in localization and SLAM, we show the relations between different estimation methodologies and state some of the advantages and disadvantages for each of them.