In this paper, we consider learning algorithm design in potential game theoretic cooperative
control. In general, standard game theory aims at just computing a Nash equilibrium and it is required for
agents' collective actions to converge to the most efficient equilibria. In particular, the equilibria maximizing
the potential function should be selected in case the utility functions are aligned to a global objective
function. In order to meet the requirement, this paper proposes a learning algorithm called Payoff-based
Inhomogeneous Partially Irrational Play (PIPIP). Finally, the effectiveness of the algorithm is demonstrated
through experiment on a sensor coverage problem.
各種検索
サポート
ヘルプ