A New Strategy for Exploring Unknown Environments for the SLAM Problem

Abstract

This article presents a new strategy aimed at solving the problem of exploration of unknown environments for SLAM, which is based in the efficient process of acquiring the maps of environments through the link between the motion planning tasks and the simultaneous localization and mapping. Although, in recent years this kind of algorithms have had a great progress, one can remember that it is a still open problem, because the solution is imposed by the human interaction that defines the destiny and the objective of the task to be carried out by the robot. Following this idea, this research project proposes a new method of deterministic exploration based on the construction of a graph as a data structure, and on the systematic exploitation of the knowledge acquired by the robot while navigating the environment.