Home /Research /Bots2ReC: Radar Localization in Low Visibility Indoor Environments
PERCEPTION

Bots2ReC: Radar Localization in Low Visibility Indoor Environments

Nils Mandischer, Sami Charaf Eddine, Mathias Huesing, Burkhard Corves

Year
2019
Citations
4

Abstract

The EU-funded project Robots To Re-Construction (Bots2ReC*) develops an operational process for the automated removal of Asbestos, employing a robotic system. In this use case, the low visibility resulting from dust formation imposes significant challenges on the environmental perception. Therefore, specialized strategies have to be developed in order to fulfill proper positioning of the mobile platform. In this paper, two strategies are introduced to navigate in enclosed environments using radar sensing, namely a novel radar SLAM and a localization strategy using laser maps. Both strategies are augmented by an efficient radar filtering process, which includes adaptive intensity thresholding, point set segmentation and hard cap noise reduction.

Keywords

RadarVisibilityComputer scienceComputer visionThresholdingProcess (computing)Artificial intelligenceMobile robotRadar imagingSegmentation

Related papers

Browse all PERCEPTION papers