A cells covering based method for Simultaneous Localization and Mapping in an unknown indoor environment

This work describes a novel solution to the mapping problem for a mobile robot moving in an unknown indoor environment. The proposed mapping technique provides a cells-based covering of the environment boundaries. No assumptions are made on the cells placement, therefore the covering is not forced to stay in a regular scheme, like in the grid-based approach. The main idea is to estimate the probability mass function of finding an environment boundary point in a given cell, and to use the obtained approximation of these points to build an environment map. The resulting mapping algorithm can be either used alone to solve the Simultaneous Localization and Mapping (SLAM) problem, or it can be used in series to an existing given SLAM algorithm to improve its performance. Numerical simulations show the effectiveness of the proposed mapping and SLAM strategies by comparing them with other SLAM techniques from the literature.