一种面向多维变体飞行器的嵌套式高效轨迹 优化方法

The solving efficiency degrades sharply due to the increased scale of control variables in the trajectory optimization of multi-dimensional morphing flight vehicles，and an efficient trajectory optimization method with a nested architecture is proposed to solve this problem. The original optimal control problem with attitude and morphing variables as control variables is equivalently transformed into a nested solving framework：an outer-layer single-dimensional optimal control problem with lift coefficient as the control variable is established，and an inner-layer static parameter optimization problem for drag minimization with morphing and attitude variables as design variables is formulated，thus realizing the equivalent dimensionality reduction solution of the original problem. The hp-adaptive Radau pseudospectral method is employed for the numerical solution of the outer-layer problem. For the inner-layer problem，an interpolation model for minimal drag coefficient optimization is constructed based on an adaptive sampling strategy，and the model is invoked during the iterative solution of the outer-layer problem to obtain the minimal drag coefficient under the specified lift coefficient. Comparative verification between the proposed nested solving framework and the direct solution method is carried out on an unpowered gliding morphing flight vehicle under different morphing dimensions. Results demonstrate that the solving time is reduced by more than 1 to 4 orders of magnitude for the optimization problems with 1 to 4 dimensional morphing variables，and the computational efficiency is independent of the problem dimensionality，which significantly improves the trajectory optimization efficiency of multi-dimensional morphing flight vehicles.