Indoor Mobile Robot Navigation in Unknown Environment Using Fuzzy Logic Based Behaviors

Abstract

This paper presents the development and design of a methodology based on fuzzy logic to control an indoor mobile robot for a complete navigation in an unknown environment. The methodology incorporates two basic behaviors, namely: reaching the goal and avoiding obstacles. The obstacle avoidance behavior is treated using wall-following scheme based on an Interval Type-2 Fuzzy Inference technique. .This helps in handling data uncertainties to produce a better performance. The mobile robot control mechanism uses some sort of knowledge base arranged in a set of fuzzy-rule-base to implement the wanted behavior that makes the mobile robot follow the boundary of an obstacle or a wall. A constant distance to the obstacle/wall is maintained while the robot tries successfully to get around this difficulty. Once the path is clear, the obstacle avoidance behavior is inhibited and reaching the goal behavior is activated using a secondary fuzzy controller. This methodology was successfully tested on a real mobile robot for different sort of scenarios. In order to provide better insight into the work's objective, a comparison work with another method, which uses a Partial Swarm Optimization-Fuzzy method, is carried out based on some defined criteria. The experiment shows a better improvement in the results of the proposed method.