【摘要】 针对变体飞行器宽域飞行引起的气动参数不确定问题，提出了一种基于回归向量动态扩张的在线参数辨识方法。首先，基于变体飞行器气动模型构建多项式回归方程，并设计广义状态观测器对回归向量进行动态扩张，将保证参数估计一致性的持续激励条件转换为区间激励条件，降低了气动参数辨识过程中激励信号设计的保守性。其次，结合扩张多项式回归方程和梯度下降算法设计参数辨识更新律，实现气动参数的在线辨识。最后，结合飞行任务场景进行仿真验证，结果表明该方法能够在区间激励条件下保证变体飞行器不确定气动参数的快速精确辨识，相较于传统辨识方法具备更强的收敛性能。更多还原

【Abstract】 Aerodynamic parameter uncertainty caused by large-envelope flight of morphing flight vehicles is addressed through an online parameter identification method based on regression vector dynamic expansion. First, a polynomial regression equation is constructed from the aerodynamic model of morphing flight vehicles, and a generalized state observer is designed to dynamically expand regression vectors. This transforms the persistent excitation condition ensuring parameter estimation consistency into an interval excitation condition, reducing conservatism in excitation signal design. Second, the parameter identification update law is formulated by integrating the expanded polynomial regression equation with a gradient descent algorithm, enabling online identification of aerodynamic parameters. Finally, the effectiveness of this approach is validated through simulation tests on flight mission scenarios. Results demonstrate rapid and precise identification of uncertain aerodynamic parameters under interval excitation conditions, with superior convergence performance compared to traditional methods.更多还原