Mobile Robot Localization: A Review of Probabilistic Map-Based Techniques

Abstract

This work presents a comprehensive review of current probabilistic developments used to calculate position by mobile robots in indoor environments. In this calculation, best known as localization, it is necessary to develop most of the tasks delegated to the mobile robot. It is then crucial that the methods used for position calculations be as precise as possible, and accurately represent the location of the robot within a given environment. The research community has devoted a considerable amount of time to provide solutions for the localization problem. Several methodologies have been proposed the most common of which are based in the Bayes rule. Other methodologies include the Kalman filter and the Monte Carlo localization filter wich will be addressed in next sections. The major contribution of this review rests in offering a wide array of techniques that researchers can choose. Therefore, method-sensor combinations and their main advantages are displayed.