Constructing Maps for Autonomous Robotics: An Introductory Conceptual Overview

Mapping the environment is a powerful technique for enabling autonomy through localization and planning in robotics. This article seeks to provide global overview of actionable map construction robotics, outlining basic problems, introducing techniques overcoming them, directing reader toward established research covering these problem solution domains more detail. Multiple levels abstraction are covered non-exhaustive vertical slice, starting with fundamental constructing metric occupancy grids Simultaneous Localization techniques. On top these, topological meshes semantic maps reviewed, comparison drawn between multiple representation formats. Furthermore, datasets metrics used performance benchmarks discussed, as challenges faced some that deviate from typical laboratory conditions. Finally, recent advances robot control without explicit touched upon.