基于人工智能深度增强学习的装备维修保障兵棋研究

Study on Equipment Maintenance and Security Based on Artificial Intelligence Depth Enhancement

为提升装备维修保障兵棋系统的学习能力和对抗水平,前沿性的将人工智能领域相关理论技术应用到系统中,通过运用马尔科夫决策过程(MDP)与神经网络等方法,在系统内部建立环境感知反馈、过程在线学习等通道,进一步扩展、增强和延伸系统中AI的角色能力,并随着推演次数的增加充分挖掘系统数据资源潜能,同步增强AI的推演行动反馈及战术策略应用能力,提升利用效率,实现推演-学习-推演的有效循环,以此来逐步提高兵棋系统的对抗推演水准,同步带动装备指挥员谋略决策能力的进一步提升,达到向实战化靠拢的要求。

In order to improve the equipment maintenance support wargame system learning ability and combat level,it applies the related theory and technique of artificial intelligence to the system. Through the use of Markoff decision process（ MDP） and neural network method, the establishment of environmental perception feedback,process of online learning channels within the system further expand,and strengthen,and extend the system the role of AI in capacity. With the increase of the number of deduction to fully tap the potential of system data resources,it enhances the synchronization deduction action AI feedback strategy and application ability, and improves efficiency, the study of effective circulation deduction,in order to gradually improve the Wargaming system level,to further drives the equipment commanders to improve their decision-making ability to meet the requirements of actual combat.