Fuzzy logic-based real-time robot navigation in unknown environment with dead ends

The proposed approach in this paper involves a new grid-based map model called “memory grid” and a new behavior-based navigation method called “minimum risk method”. The memory grid map records not only the environmental information, but also the robot experience. The minimum risk method is just one of the applications of the memory grid technique, which addresses the local minimum problem faced by a goal-oriented robot navigating in unknown indoor environments. The Minimum Risk implies that the robot is able to choose the safest region that can avoid colliding with obstacles and prevent the robot from iterating previous trajectory. This method is demonstrated to work in long wall, large concave, recursive UU-shaped, unstructured, cluttered, maze-like, and dynamic indoor environments. It adopts a strategy of multi-behavior coordination in which a novel path-searching behavior is developed to recommend the region offering the minimum risk. Fuzzy logic is used to implement the behavior design and coordination. The proposed approach is verified with simulation and real-world tests.