Biologically inspired robot swarm control for subaqueous environment

Abstract

The goal of this work is to conceptualize a robust and flexible behavior controller not only for basic swarming behavior of homogeneous underwater robot teams, but also for advanced missions in subaqueous environment based on robust communication. Therefore, this paper proposes a modularized control architecture that represents the imitation of dolphin's phylogenetic species intelligence such as sensing and interpreting abilities and swarming habits of dolphins, but also additional skills from the learned intelligence. The swarming represents the basic behavior given in the natural environment. The mission consisting of behavior primitives can be structured by reinforcement learning to perform advanced missions. A knowledge based sensing module can be implemented additionally to improve the communication robustness. This paper especially focuses on the knowledge based communication concept. © 2011 IEEE.