Behavior Based Robot Navigation: Resolving Behavior Conflicts Using Fuzzy Inference System

摘要

Conflict resolution is the control decision process, which should be taken as a result of the firing among several fuzzy behavior rules. In the Behavior-based Robot Navigation System, control of a robot is shared between a set of perception-action units, called behaviors selection. In other words, the behavior selection is the way that an agent selects the most appropriate or the most relevant next action to take at a particular moment, when facing a particular problem. Based on selective sensory information, each behavior produces immediate reaction to control the robot with respect to a particular objective, i.e., a narrow aspect of the robot’s overall task such as obstacle avoidance or goal seek. Behaviors with different and possibly incommensurable objectives may produce conflicting actions that are seemingly irreconcilable. The main issue in the design of behavior based robot control systems is the formulation of effective mechanism to coordinate the behavior’s activities without any behavior conflicts during navigation. This paper presents the techniques to design the behaviors and resolve the behaviors conflicts, which are based on the Situation Context of Applicability (SCA) of the environments.