面向变形滑翔飞行器的智能变形决策方法

打开文本图片集

关键词：变形滑翔飞行器；智能变形决策；深度强化学习

中图分类号：V448文献标志码：A

DOI： 10.7652/xjtuxb202605021 文章编号：0253-987X（2026）05-0217-09

Intelligent Deformation Decision-Making Method for Morphing Glide Vehicle

WANG Taojun，MENG Fanyi， CHEN Gang （School of Aerospace Engineering，Xi'an Jiaotong University，Xi'an 710o49，China）

Abstract： To fully leverage the significant advantages of morphing glide vehicles （MGVs） over traditional fixed-configuration aircraft in terms of range，speed，and environmental adaptability， and to address the challenge of optimal morphing configuration decision-making under multiphysics coupling，an intelligent morphing decision-making method based on deep reinforcement learning is proposed. First，a deep neural network was used to construct a surrogate model for wing surface morphing and aerodynamic performance，achieving end-to-end rapid mapping from morphing actions to aerodynamic data. Then，a reinforcement learning framework was established by integrating the longitudinal motion model of the vehicle. Through comparative analysis of algorithm performance，the proximal policy optimization （PPO） algorithm was selected to construct the inteligent decision-making model，enabling autonomous policy learning under aerodynamic/morphing/trajectory cross-coupling. Finally，full-process simulation experiments were conducted in perturbed environments beyond the training scope for validation. The results show that the proposed method can inteligently make morphing decisions by comprehensively considering the coupling effects of aerodynamics，morphing，and trajectory，increasing the terminal range of the MGV by 7.304% and effectively maintaining a high lift-to-drag ratio flight state. Additionally， perturbation tests demonstrate that the method remains feasible and exhibits strong generalization capabilities in uncertain environments.

Keywords： morphing glide vehicle； intelligent deformation decision-making；deep reinforcement learning

随着科学技术的发展和飞行任务需求的日趋复杂化，固定构型飞行器有限的任务执行区间和执行能力，严重制约了飞行器生存能力和实际飞行效能的提升，因此亟需发展先进的气动布局及控制方式。（剩余12585字）