Mobile Robot's Map Reconstruction Based on DSmT and Fast-Hough Self-Localization

This paper presents a new tool based on DSmT (Dezert-Smarandache Theory) and fast-Hough self-localization which is applied to autonomous mobile robot's map reconstruction. Uncertainty from sonar information is fused and managed in rebuilding grid map effectively with DSmT, since fast-Hough self-localization from laser information overcomes some shortcomings of classical Hough transformation including high amount of computation and low precision of localization. When a virtual mobile robot evolves in the virtual environment with different outline features, the result can adequately testify that the new tool owns better performance to rebuild map. In general, this research provides an approach for studying autonomous navigation of mobile robot or unmanned land vehicle (ULV) in unknown environment, also provides a human- computer interactive interface to manage and manipulate the robot remotely.