A review and comparison of mapping and trajectory selection algorithms

Abstract

The dire need to solve orientation and localization tasks is directly related to the development of autonomous robotics systems as autonomous modules. In this article, we have reviewed and analyzed possible areas and peculiarities when implementing existing localization approaches in autonomous robotics systems operating under various weather conditions with possible obstacles on their way without preliminarily generated maps. In the paper, we especially pay attention to existing SLAM algorithms and a multitude of hardware concerned with this problem. Every considered and addressed algorithm in this paper comes with its main principles and generated, as a result of its performance, map type. The comparison of the algorithms was mainly based on the data of several articles and projects, in which almost perfect indoor experiments without any weather impact in order to examine the efficiency of the algorithms were conducted. Using the results acquired by the authors, a comparative table with main statistics for every considered algorithm was created. Apart from that, similar statistics for trajectory selection algorithms that meant to help researchers solve scenario/scripted tasks were covered. As a result of our review piece, we presented a ranging technique for the pair algorithms/sensors that uses the renowned TOPSIS outranking methodology. The proposed approach may become of significant help while selecting the pair for every case study.