Behavior Based Mobile Robot Navigation in a Dynamic Environment: Fuzzy Logic Approach
Priyanka Bansal, Sharad Maheshwari, Manish Kumar and Umesh Kumar
This paper proposes a fuzzy logic based behavior control for a mobile robot in a dynamic environment. The control problem can be broken down into smaller units of behavior. The navigation system in this paper consists of two behaviors – the obstacle avoidance behavior and the target seeking behavior. One of the problems associated with traditional behavior based navigation systems for mobile robots are that of behavior coordination; how to coordinate the simultaneous activities of several independent behavior producing units. The main idea of this paper is to coordinate their conflicts using fuzzy sets and fuzzy rules. Weights are assigned to each behavior using fuzzy logic (14 rules are used for assigning weights) and then weighted average of crisp values is taken to produce final output. The robot is modelled in Matlab and the fuzzy logic rules are optimized for the best results possible. The simulation results show that the proposed method can perform robot navigation in uncertain and dynamic environments.
Keywords: Collision avoidance, path planning, robot navigation.