Behavior-Based Blind Goal-Oriented Robot Navigation by Fuzzy Logic

This paper proposes a new so-called “minimum risk” method, to address the local minimum problem that has to be faced for the goal-oriented robot navigation in an unknown environment. This method is theoretically proved to guarantee the global convergence even in the long-wall, large concave, recursive U-shape, unstructured, cluttered, and maze-like environments. The minimum risk method adopts a strategy of multiple behaviors coordination, in which a novel path-searching behavior is developed by fuzzy logic to recommend the regional direction with minimum risk. This behavior is one of the applications of the proposed memory grid technique. The proposed method is verified by the simulation and real world tests.