提出了一种基于单步预测的多级影响图法,并将该理论应用在单机连续空战机动决策系统建模中。在决策系统的建模过程中考虑了影响无人机机动决策的各要素,包括无人机动力学方程、决策者的偏好以及战斗双方态势的不确定性等。机动决策的态势基准是通过无人机动力学方程迭代获得。运动方程中的控制量通过求解单步预测影响图获得。该预测影响图算法,效能函数的起点不断向后平移,建立了离散空间空战决策的纳什平衡,通过求解效能函数的最大值就能获得每一步的最优控制解。最后仿真验证该方法有效性,无人机模型采用质点运动方程,使用VC++实现并机动决策算法。分析表明提供了一种新型无人机机动决策算法,发展了无人机自主空战机动决策系统。
A multistage influence diagram game based on one-step prediction is put forward for modeling the maneuvering decisions of UCAV in air combat. In modeling the decision system, the elements that have effect on UCAV maneuvering decision are taken into consideration, including the dynamical equation of UCAV, the pilots'preferences and conditions of uncertainty and so on. The UCAV optimal control sequences with respect to dynamical models are obtained by solving the influence diagram game with a one-step prediction control approach. In this approach, the time horizon of the original game is truncated, and a feedback Nash equilibrium of the dynamic game is determined and implemented at each decision stage. To demonstrate the validity of the influence diagram game, particle motion equation was used in UCAV modeling, and VC ++ presented here offers a was used to implement the maneuvering decision. Analysis showed that the method new UCAV maneuvering decision algorithm, and enhances UCAV autonomous decision-making system.