Coordinating Heterogeneous Swarms through Minimal Communication among Homogeneous Sub-swarms

robotics, the agents are often assumed to be identical. In this abstract, we argue that the cooperation between swarms of different kinds of robots can enhance the capabilities of the robotic system--heterogeneous swarms marry the robustness and parallelism of homogeneous swarms with efficient task specialisation. A key issue in heterogeneous swarm systems is the potential complexity of facilitating cooperation between the different robot types. We propose an approach to heterogeneous system design that minimises the complexity of heterogeneous interaction whilst preserving the benefits of specialisation. To mitigate this complexity, we restrict interactions between homogeneous sub-swarms to (very) simple forms of communication. This restriction allows the system to be completely modular. At the top level, modules are globallevel behaviours executed by the heterogeneous robotic swarm. Each global-level behaviour is obtained by decomposing the heterogeneous swarm into its homogeneous co...

• Type: Journal
• Year: 2010
• Where: ANTSW
• Authors: Carlo Pinciroli, Rehan O'Grady, Anders Lyhne Christensen, Marco Dorigo
• Keywords: ANTSW 2010, Artificial Intelligence, Heterogeneous Swarm, Robotic, Swarm