engleski

Histogram Based Mobile Robot Localization

Global localization refers to the problem of determining the pose of a mobile robot under global uncertainty. One solution to this problem is the usage of histograms as part of a topological map. The algorithm introduced here is based on the interacting multiple model and exploits a soft gating of the problem to reduce the computational requirements of presented approach. This localization algorithm consists of a position estimation and orientation estimation part. The position part is based on a xy histogram scan matching procedure, where xy histograms are extracted directly from local occupancy grid maps. The orientation part is based on the proposed obstacle vector transform combined with polar histograms. Introduced algorithm is tested using a Pioneer 2DX mobile robot simulator.

mobile robots; localization; histograms

nije evidentirano