Mapping the Landscape of SLAM Research: A Review

Abstract

Multiple publications have arisen from over three decades of research in the field of simultaneous localization and mapping (SLAM), overwhelming those who wish to delve into this area. The extensive body of work in SLAM has resulted in an abundant and, at times, confusing flow of data, lacking a straightforward explanation of the underlying principles. This article aims to address this issue by providing a clear overview of the general taxonomy of the SLAM universe, assuming the reader is completely unfamiliar with the subject area. As cameras remain the primary sensor choice for SLAM, and considering the vast number of articles available on this topic, special emphasis will be placed on Visual SLAM. Additionally, we will delve into the influence of artificial intelligence on the development of new algorithms. The article incorporates comparative tables to enable readers to assess system performance using benchmark datasets. Moreover, it offers insights into current trends and prospective pathways within the subject area.