Multiagent Multiobjective Decision Making and Game for Saving Public Resources.

Uncertain environments and inefficient decision analysis restrict the efficient utilization of depletable public resources by multi-agents, especially for the scenario involved with multi-objective game dilemmas and weak scalability of decision-making. To address above conundrums, this paper proposes a multi-layer games framework that integrates cognition, decision-making, and countermeasures (CDC). Through the transformation of agent preference to alliance communication structure, a cooperation-competition topology network (CCTN) model is constructed, which improves the convergence and solution efficiency of the game model. In view of the Gaussian kernel ascending dimension mapping, a game equilibrium particle swarm optimization (GEPSO) algorithm is designed to improve the efficiency of finding equilibrium solutions and solve the non-deterministic polynomial (NP) problem of multi-objective games. To validate the effectiveness and performance of the proposed methodology, a case study of collaborative detection of multi-vehicle is conducted using the proposed framework and model.