高超滑翔飞行器对地侦察轨迹的分段优化策略  

Segmented optimization strategy for ground reconnaissance trajectory of hypersonic glide vehicle

作　　者：王金硕 何冉 刘海平 Wang Jinshuo;He Ran;Liu Haiping(Beijing System Design Institute of Electro-Mechanic Engineering,Beijing 100854,China)

机构地区：[1]北京机电工程总体设计部,北京100854

出　　处：《战术导弹技术》2025年第1期113-125,135,共14页Tactical Missile Technology

摘　　要：针对在高超声速滑翔飞行器对地侦察的轨迹优化中,因目标函数形式复杂导致传统优化方法易陷入局部最优的问题,提出了一种自适应Radau伪谱法与近端优化策略算法相结合的分段优化策略。将全程轨迹划分为滑翔抵近段和目标侦察段,并预先设计轨迹交接点方案,确保阶段间状态的平滑过渡,同时使得飞行器状态满足侦察任务需求。利用自适应Radau伪谱法和近端优化策略分别对滑翔抵近段和目标侦察段进行优化,结合两种方法的高效性和模型普适性,实现快速、精确求解。数值仿真证明,所提分段优化策略可以高效解决具有非常规目标函数的轨迹优化问题,为此类研究提供新的视角。In trajectory optimization for ground reconnaissance by hypersonic glide vehicles, the presence of the complex objective function can render traditional optimization methods prone to converging on local optima. Aiming at this problem, a segmented optimization strategy combining the adaptive Radau pseudospectral method with the proximal optimization strategy is proposed. The entire trajectory is divided into a glide approach segment and a target reconnaissance segment, and the transition point scheme is pre-designed. This scheme ensures the smooth transition between two phases while ensuring that the vehicle′s states can fulfill the reconnaissance mission requirements. The adaptive Radau pseudo-spectral method and proximal optimization strategy are respectively employed to optimize the glide approach segment and target reconnaissance segment.By combining the high efficiency and model applicability of these two approaches, the rapid and accurate solution can be guaranteed. Numerical simulations demonstrate that the proposed segmented optimization strategy can effectively solve trajectory optimization problems with unconventional objective functions, providing a new perspective for similar studies.

关 键 词：高超声速滑翔飞行器 轨迹优化 最优控制 自适应伪谱法 深度强化学习 对地侦察 蒙特卡洛 

分 类 号：V448[航空宇航科学与技术—飞行器设计]