Audio communication for multi-robot systems

Interaction through communication is an important aspect of multi-robot systems. Audio communication, while common among animals and well studied in biology, is not well explored as a multi-robot technology. In this thesis we study the use of audio messages as a means of communication among mobile robots. We examine the properties of audio compared to other communication media, and show how these can be exploited. To guide the design of a multi-robot system, a simple audio propagation model integrated into a robot simulator is developed. This simulator shows how acoustic communication improves the team performance in a prototypical search task. We also introduce a physical network layer that uses audio as the transmission medium and implements a broadcast method related to the CSMA protocol. We then describe a distributed mutual exclusion algorithm suitable for use over audio, and demonstrate it in both simulation and real-world.