【摘要】 针对高超声速变形飞行器的时间协同再入滑翔制导问题，提出了伪谱在线规划协同制导算法。首先，建立了考虑乘波体底部出口变形和地球自转影响的降阶滑翔动力学模型，该模型对飞行器经度、纬度、航向角和飞行时间有很高的预测精度，可将积分计算速度提高一个数量级；其次，为降低数值求解规模，将控制量参数化为倾侧角余弦、两个倾侧翻转点能量和外形参数，并将偏差传播方程线性化；然后，使用支撑点为LG点的Lagrange插值多项式对状态和控制量进行离散，使用伪谱方案，在多项式空间内近似推导出控制的解析修正公式，成功建立多段线性高斯伪谱时间协同模型预测控制算法，可同时进行在线弹道规划与变形决策；最后，标称和蒙特卡洛打靶的仿真结果均表明，该算法具有控制精度高、计算速度快、适应性好、鲁棒性强的优势。更多还原

【Abstract】 To address the time-coordinated reentry glide guidance problem for hypersonic morphing vehicles, a pseudospectral online planning coordinated guidance algorithm is proposed. First, a reduced-order glide dynamics model is established, considering the deformation of waverider’s bottom exit and the Earth’s rotation effects. This model achieves high prediction accuracy for the vehicle’s longitude, latitude, heading angle and flight time while improving the computational speed of integral calculations by an order of magnitude. Second, to reduce the complexity of numerical solutions, control variables are parameterized as the cosine of the bank angle, energies at two bank reversal points and shape parameters. The deviation propagation equations are then linearized. Next, the states and controls are discretized using Lagrange interpolation polynomials with Legendre-Gauss nodes. A pseudospectral scheme is employed to approximate solutions in polynomial space, leading to the derivation of analytical correction formulas for control updates. This approach successfully establishes a multi-phase linear Gaussian pseudospectral time-coordinated model predictive control algorithm that enables simultaneous online trajectory planning and deformation decision-making. Finally, both nominal and Monte Carlo simulation results demonstrate the algorithm’s advantages in terms of high control accuracy, fast computation speed, strong adaptability and robustness.更多还原