一种高速可变形飞行器智能变形决策方法

针对一类高速可变形飞行器(HMFV)的变形决策问题，提出一种基于深度确定性策略算法(DDPG)下考虑综合性能指标最优的智能变形决策方法。首先，以一类后掠角可连续变化的高速飞行器为研究对象，给出变形飞行器动力学模型，分析模型特性及变形量与关键气动参数之间的定性关系。其次，基于关键气动数据特征分析，考虑包含气动性能、控制误差在内的综合性能指标，设计一种基于DDPG算法的智能变形决策方案。再者，针对带有标称控制器的HMFV进行变形决策训练，实时获得滑翔过程中不同飞行状态下的最优构型。最后，仿真结果表明所设计的智能变形决策算法收敛效果好，且具备较好的泛化性能。相比于固定外形，可通过变形使得在不同状态下的升阻比保持最优，且与考虑单一决策指标相比，考虑综合指标最优的变形决策可进一步缩小姿态动态跟踪误差。

An Intelligent Deformation Decision making Method for High speed Morphing Flight Vehicle

To address the deformation decision making problem for a class of the high speed morphing flight vehicle (HMFV), an intelligent deformation decision making method based on deep deterministic policy gradient (DDPG) is proposed by considering the optimal comprehensive performance index. Firstly, the qualitative relationship between deformation and aerodynamic characteristics is analyzed for the variable sweptback high speed flight vehicle. Secondly, a comprehensive performance index including lift to drag ratio and attitude control error is designed, and an intelligent decision method based on DDPG algorithm is introduced. Thirdly, to obtain the optimal configuration in the gliding phase, the deformation decision making training is carried out with HMFV nominal controller. Finally, the numerical simulations demonstrate that the designed intelligent deformation decision algorithm converges well and has good generalization performance. Compared with the fixed shape, the lift to drag ratio can be kept optimal under different states by deforming the sweptback angle, and the deformation decision under the optimal comprehensive index can further reduce the dynamic attitude tracking error.