SLAM for autonomous planetary rovers with global localization

• DOI: 10.1002/rob.21943
• Publication Date: 2020-02-29T05:00:20Z
• Authors: Dimitrios Geromic...; Martin Azkarate; Emmanouil Tsardou...; Levin Gerdes; Loukas Petrou; Carlos Perez Del ...
• Keywords: Orbiter

Abstract This paper describes a novel approach to simultaneous localization and mapping (SLAM) techniques applied the autonomous planetary rover exploration scenario reduce both relative absolute errors, using two well‐proven techniques: particle filters scan matching. Continuous is improved by matching high‐resolution sensor scans online created local map. Additionally, avoid issues with drifting localization, globally corrected at discrete times, according predefined event criteria, current map orbiter's global The resolutions of maps can be appropriately chosen for computation accuracy purposes. Further, generated map, form structured elevation grid also used evaluate traversability surrounding environment allow continuous navigation. objective this study support long‐range low‐supervision exploration. implemented SLAM technique has been validated data set acquired during field test campaign performed Teide Volcano on island Tenerife, representative Mars/Moon scenario.