论文部分内容阅读
研究非固定时间的航天器双脉冲交会轨迹优化问题,设计了基于梯度分割区间优化算法(GIOA)。该算法结合所研究问题的特点,使用每次只选择有限个区间进行操作的区间选择策略、基于梯度优化结果的区间分割策略、基于单调性的区间紧缩策略以及约束条件测试和基于梯度的目标优化估计值更新策略等。梯度优化算法仅用于区间分割和目标优化估计值更新,不但没有影响GIOA对区间优化算法全局性和收敛性的继承,同时加快了包含优化解的小宽度区间的出现,提高了目标优化估计值的更新速度,并由此提高了运算效率。区间选择策略的使用,控制了决策变量区间数量的增长,降低了算法运行的存储需求。算例仿真中,成功求解非固定时间双脉冲交会问题,并展示出算法的优势。
The non-stationary spacecraft dual-pulse rendezvous trajectory optimization problem is studied, and the gradient partition interval optimization algorithm (GIOA) is designed. The algorithm combines the characteristics of the studied problems with the interval selection strategy that only chooses a limited number of intervals at a time, the interval segmentation strategy based on gradient optimization results, the interval tightening strategy based on monotonicity, the constraint test and the gradient-based target optimization Estimation update strategy and so on. The gradient optimization algorithm is only used for interval segmentation and target-optimized estimation update, which not only does not affect the GIOA inheritance of the global and convergence of the interval optimization algorithm, but also accelerates the appearance of small-width interval containing the optimal solution and improves the target optimization estimation The update rate, and thus improve the computational efficiency. The use of interval selection strategies controls the growth of the number of decision variables and reduces the storage requirements for algorithm operation. In the example simulation, the non-fixed time double pulse rendezvous is successfully solved and the advantages of the algorithm are demonstrated.