基于确定性梯度策略的固定翼飞行器博弈方法

Fixed-wing Aircraft Game Method Based on Deterministic Policy Gradient

飞行器博弈对抗是无人系统领域的重要研究方向,而传统的控制方法在这一领域中存在着不稳定、精度低以及临机决策响应速度慢等问题。为了解决这些挑战,充分利用大数据和数据驱动的优势,提出了一种基于深度确定性策略梯度算法(DDPG)的数据驱动控制方法,通过大规模数据的训练和优化,增强飞行器的控制能力,帮助飞行器在复杂环境下实现自主决策。搭建仿真环境,对所提出的算法进行验证。实验结果充分证明数据驱动方法在飞行器博弈对抗中的可行性和有效性。

Aircraft game confrontation is an important research direction in the field of unmanned systems.However,the traditional control methods in this field have problems such as instability,low precision,and slow response speed of ad hoc decision-making.To address these challenges and take full advantage of big data and data-driven advantages,this paper proposes a data-driven control method based on the deep deterministic policy gradient algorithm(DDPG),which enhances the control ability of the aircraft through training and optimization on large-scale data to help aircraft achieve autonomous decision-making in complex environments.Finally,a simulation environment is built to verify the proposed algorithm.The experimental results fully prove the feasibility and effectiveness of the data-driven method in the game confrontation of aircraft.