PROJECT TITLE :

Topological Indoor Localization and Navigation for Autonomous Mobile Robot

ABSTRACT:

Mobile robot typically has limited on-board resources and could be applied in different indoor atmosphere. Therefore, it is necessary that they will learn a map and navigate themselves autonomously with lightweight algorithms. A novel topological map-building-primarily based localization and navigation methodology is proposed during this paper. Primarily based on the depth curve provided by a 3D sensor, a progressive Bayesian classifier is developed to understand direct corridor type identification. Instead of extracting options from single observation, data from multi-observations are fused to realize a a lot of robust performance. A topological map generation and loop closing methodology are proposed to make the surroundings map through autonomous exploration. Based mostly on the derived map and therefore the Markov localization methodology, the robot will then localize itself and navigate freely in the indoor environment. Experiments are performed on a recently designed mobile robot system, and therefore the results verify the effectiveness of the proposed methodology.