基于遗传算法的RLV再入轨迹优化设计

Genetic algorithm optimization of RLV reentry trajectory

飞行器再入轨迹优化是一类最优控制问题。传统的优化方法存在初始值敏感问题。利用小生境技术和精英方法改进适应值共享拥挤遗传算法,并将其用于RLV再入轨迹优化设计。以终端时间固定的空间最小控制能量再入轨迹和终端时间自由的平面最小热载再入轨迹为例,详细讨论了遗传算法用于再入轨迹优化设计所需要解决的一些关键问题。仿真结果表明提出的方法能够较快地搜索到全局最优解,对初始猜测值不敏感,能够方便用于RLV的再入轨迹方案选择和优化设计。

Spacecraft trajectory optimization is a kind of optimal controt program.The conventional direct and indirect methods are all very sensitive to the initial guess that needs much experience. The improved float point algorithm and fitness sharing genetic algorithms （FSGA） with elitism method both are applied to solve this problem. For the minimum control energy reentry trajectory with fixed terminal time and the minimum total heat load reentry trajectory, some key problems such as parameterized method are discussed in detail. The simulation results indicate that the global optimization results can be caught faster and is not sensitive to the initial guess. The improved genetic algorithm is easy to use in RLV reentry concept design and trajectory optimization.