Skip to Main Content
In consideration of the diversity of ship mission, uncertainty of the environment and the requirements for the intelligent management of marine power platform, the technology of multi-agent is introduced into the autonomous operation of shipboard fluid system and the hierarchical architecture of multi-agent intelligent control is presented. The method of optimal resource allocation with the existence of uncertainty is analyzed. Firstly, the resource allocation problem is formulated as a loosely coupled Markov decision process (MDP) and the multi-agent MDP model is established. Then the optimal resource allocation is achieved by following the optimal policy which is obtained by using genetic algorithm (GA) based on the model. Lastly, the performance of the optimal policy is examined in the emulation. The research can afford some academic guidance for raising the level of intelligence, reducing the manpower for ship operation, and increasing the mission effectiveness.