战场迷雾中的欺骗路径规划

Deceptive Path Planning in Fog of War

计算机生成兵力是由计算机生成的虚拟作战力量对象,是军事仿真领域的关键要素。欺骗路径规划是欺骗行为的一种基本呈现模式,对提高计算机生成兵力的智能性和竞争力具有重要意义。然而,目前欺骗行为与军事仿真的结合还不够充分,经典的路径规划方法不能有效利用战场的部分可观测性,进而达到更好的欺骗效果。为了解决这些问题,在路网定义的基础上重新定义了一种,并基于四种经典的欺骗策略,提出了四种新的欺骗路径规划方法。为了对新方法和经典方法进行可视化比较,在基于KD-HLA-RTI中间件搭建的坦克对抗实验平台上进行了红蓝对抗实验。红色坦克作为被观察者,依据不同的策略进行欺骗路径规划,而蓝方作为观察者,依据观测序列对红色坦克的真实目标进行在线识别,并为其部署防御资源。实验结果表明,在战场环境部分可观测条件下,与经典方法相比,四种新方法能在确保更强欺骗性的同时,更有效地减少被观察者抵达真目标的时间。后续多次仿真实验表明,第三种方法在欺骗效果的提升上对于战场环境变化有较好的鲁棒性。

Computer generated forces(CGFs)are virtual combat force objects created by computers and critical elements in the field of military simulation.Deceptive path planning is a basic method of deceptive behavior,which is important for improving the intelligence and competitiveness of CGFs.However,the current combination of deceptive behavior and military simulation is insufficient,and classical path planning methods cannot effectively take advantage of the partial observability of the battlefield and achieve better deceptive effects.To solve these problems,we propose four new deceptive path planning methods by re-defining a single circular fog road network based on road networks and four classical deceptive strategies.In order to make a visual comparison between the new and classical methods,we conduct a red-blue adversarial scenario experiment on the KD-HLA-RTI-based tank combat platform.The red tank plays the role of the observed and conducts path planning according to different strategies,while the blue side,as the observer,identifies the real goal of the red tank in real time and deploys defense resources to it.The experimental results indicate that compared with the classical methods,the new ones can improve the time efficiency and deceptive effectiveness more effectively for the observed under partially observable conditions of the battlefield environment.Subsequent multiple simulation experiments demonstrate that the third approach exhibits better robustness to battlefield environmental changes in terms of improving the deceptive effects.