An Automatic Mesh Refinement Method for Aircraft Trajectory Optimization Problems
An Automatic Mesh Refinement Method for Aircraft Trajectory Optimization Problems
复制标题
DOI：
10.2514/6.2013-4555
复制
发表时间：
2013-08
期刊：
影响因子：
--
通讯作者：
M. Bittner;Philip Bruhs;M. Richter;F. Holzapfel
中科院分区：
文献类型：
--
作者：
M. Bittner;Philip Bruhs;M. Richter;F. Holzapfel
针对飞行器轨迹优化问题，提出了一种控制网格自动加密算法。在手头的文件中，直接多次射击的方法，但该程序也可以与其他离散化方法相结合。该算法考虑到所考虑的动态系统的特殊性质。它是基于密度函数的思想，这也被用于DENMRA算法（参考文献1）。与只考虑控制历史来计算密度函数而不考虑底层系统的信息的DENMRA相反，这里使用动态系统的状态以及控制。因此，所得到的密度函数是专门为飞机动力学量身定做的，但由于对状态历史有很好的了解，因此可以更精确地计算。更好的代表性的控制和更好的收敛性的整体问题证明在一个例子中考虑一个简单的空中比赛的情况。
An automatic grid refinement algorithm for the control grid within an aircraft trajectory optimization problem is presented. In the paper at hand, a direct multiple shooting approach is used but the procedure can also be applied in combination with other discretization methods. The algorithm takes the special properties of the considered dynamic system into account. It is based on the idea of a density function which is also used in the DENMRA algorithm (Ref. 1). In contrast to DENMRA which is only considering the control histories to calculate the density function but no information about the underlying system, here, the states of the dynamic system as well as the controls are used. Hence, the resulting density function is specially tailored to aircraft dynamics but can be calculated more precisely because of the good knowledge of the state histories. The better representation of the controls and the better convergence of the overall problem is demonstrated in an example considering a simple air race scenario.