Navigational task-planning of distributed cooperative robotic vehicles

Real time obstacle avoidance is one of the intelligent tasks for intelligent mobile robots. Mobile robots should possess the ability to act autonomously in the presence of uncertainty and to adjust their action based on sensed information. They also should be capable of accepting high level mission oriented commands, integrate several kinds of data, including task specification, able to handle information about their own state and the state of the environment too, and be capable of reasoning under uncertainty without human intervention. The proposed technique is a behavior-based approach that blends subset navigational behaviors such as reflexive, potential field and wall following techniques. We discuss each approach separately and present a technique for adaptive navigational behavior switching that is conditional based on the availability of sensory information locally and globally. Our navigation control strategies are developed fully in the FMCell environment. We present some simulation results from FMCell software.