Development of multiobjective trajectory-optimization method and its application to improve aircraft landing
Development of multiobjective trajectory-optimization method and its application to improve aircraft landing
复制标题
DOI：
10.1016/j.ast.2016.08.019
复制
发表时间：
2016-11
影响因子：
5.6
通讯作者：
N. Othman;Masahiro Kanazaki
中科院分区：
文献类型：
--
作者：
N. Othman;Masahiro Kanazaki
本研究的目的是开发一种利用有效的时间序列飞行模拟的轨迹优化方法。通过估算时间序列空气动力学数据，求解了运动方程。根据气动数据库对任意飞机的空气动力学进行估计，利用克里格方法提高了计算效率。为了提高气动数据库的精度，需要获取额外的数据集。将该方法应用于着陆进近轨迹优化的多目标问题。考虑了两个目标函数：代价函数的最小化（代价函数表示最优轮廓轨迹）和最大加速度的最小化。采用基于Kriging模型的非支配排序遗传算法- ii作为优化器。根据结果，以及对有和没有微爆裂效应的情况的比较，微爆裂效应可以通过最小化成本函数和最大加速度来潜在地引起对飞机轨迹的高估。因此，需要将弹道修正为不产生微爆效应、具有最佳控制角度和高度的最近轨迹。气动数据库的初始采样和附加采样对轨迹优化有影响。研究表明，该数据库对于在运动方程的基础上优化轨迹分析具有重要意义。
The objective of this study is to develop a trajectory-optimization method using an efficient time-series flight simulation. Equations of motion (EoMs) were solved by estimating the time-series aerodynamics data construction. The aerodynamics of an arbitrary aircraft were estimated according to an aerodynamic database, which improved the efficiency via the Kriging method. To increase the accuracies of aerodynamic databases, additional data sets were acquired. The developed method was applied to the multi-objective problem of trajectory optimization for landing approaches. Two objective functions were considered: the minimization of the cost function, which indicates the optimal profile trajectory, and the minimization of the maximum acceleration. A Kriging model-based exploration with non-dominated sorting genetic algorithm-II was used as an optimizer. According to the results, as well as a comparison of the cases with and without the microburst effect, the microburst effect can potentially cause an overestimation of the aircraft trajectory by minimizing both the cost function and the maximum acceleration. The trajectory thus needs to be corrected to the nearest trajectory without a microburst effect, with optimal control angle and altitude. The trajectory optimization was affected by the initial sampling and additional samples of the aerodynamic database. This study shows the importance of this database for optimizing the trajectory analysis on the basis of the equation of motion.