(Georgia Institute of Technology, 2003)
Balch, Tucker; Ulam, Patrick D.
We present a means in which individual members of a multi-robot team may allocate themselves into specialist and generalist niches in a multi-foraging task where there may exist a cost for generalist strategies. Through the use of reinforcement learning, we show that the members can allocate themselves into effective distributions consistent with those distributions predicted by optimal foraging theory. These distributions are established without prior knowledge of the environment, without direct communication between team members, and with minimal state.