Evaluation of a Novel Radar Based Scanning Method

Abstract EN

The following paper introduces a novel scanningmethod for mapping and localization purposes in mobile robotics.

Our method is based on a rotating monostatic radar network,which determines the positions of objects around the scanner viaa continuously running lateration algorithm.

The estimation ofsurfaces with ultrawideband radar networks has been studiedexperimentally in lab environments, especially with lateration,envelopes of spheres, and SEABED algorithms.

But we do notsee a link to the field of mapping and localization of mobilerobots, where laser scanners are dominating.

Indeed, only fewresearch groups use radars for mapping and localization, buttheir applied sensor principle is based on a rotating focused radarbeam.

Consequently, only 2D radar scanners are known insidethe robotic world and methods for 3D scanning with radars needto be investigated.

This paper will derive the theoretical background of thesensor principle, which is based on a radar network on a rotatingjoint, and discuss its erroneous influences.

We were performingfirst scans of standard geometries and deriving a model inorder to compare theoretical and experimental measurementresults.

Furthermore, we present first mapping approaches anda simulation of a scanner with multiple sensors.