关键词: 兵棋推演, 智能体, 作战方案, 深层神经网络, 强化学习, 模仿学习

Abstract:

To enable the agent to cope with complex battle scenarios and objectives in wargame, a learnable wargame agent architecture driven by a battle scheme is proposed. By analyzing the "attachment characteristics" and "loose coupling characteristics" of the agent to wargame system, the learnable requirements of the agent are obtained. In the design of the agent framework, battle schemes are used to reduce the learning range of the agent. The finite state machine corresponds to the knowledge of the operational phase in the battle scheme, and the decision-making space of the agent is determined according to the framework of the battle scheme. A learnable deep neural network is designed to explore key decision space. The neural network uses prior knowledge imitation learning mode and deep reinforcement learning mode. This architecture can iteratively explore optimal deployment and collaboration issues for multiple chessmen that are difficult for humans to fully tease out.

Key words: wargame, agent, battle scheme, deep neural network, reinforcement learning, imitation learning