Towards Bio-inspired solutions for underwater multi-robot communication and exploration

n an underwater environment, a multi-robot exploration application faces many challenges. First, deploying many underwater robots working concurrently causes communication interferences which can jam the information sharing among robots. Second, the underwater sensing and communication range are very limited compared to surface or air robotic systems. Third, the reliable underwater odometrical measurement required for navigation needs a complex and precise model. This thesis contributes to the research field by formulating novel solutions for the underwater multi-robot communication and exploration problems. The proposed approaches are inspired from nature, by considering that simple and scalable solutions can be discovered from the results of millions of years of evolution. As an addition, some practical research work to develop underwater robot platforms, especially concerning electronic design, is also presented. The problems concerning the limited communication fidelity in an underwater multi-robot system are solved by integrating a multi-modal bio-inspired communication system and a novel scheduling algorithm. The solution is simple and decentralized, so it is scalable and feasible for an underwater swarm with many members (tens to hundreds of robots). Furthermore, some collective-exploration algorithms inspired from animal foraging behaviors are also investigated in this thesis. Both approaches are investigated by using 3D underwater swarm simulator and real robots.