面向多星区域观测调度的改进型自适应遗传算法

针对传统优化算法在解决多星区域观测调度问题中收敛速度缓慢和易于陷入局部最优解的不足，提出了一种改进型自适应遗传算法。该算法通过蒙特卡洛方法结合Hamming距离，给出较优的初始种群；根据种群的平均Hamming距离确定交叉和变异操作的执行顺序，并结合sigmoid函数和高斯函数基于种群的个体适应度设计了自适应非线性的交叉率和变异率；结合双精英保留策略和锦标赛策略，保证最优个体的遗传；使用双重停机条件，提高算法的搜索效率。最后，通过实验表明，该方法可以显著提高全局搜索能力，加快算法的收敛速度，有效提高卫星的观测效率。

An improved adaptive genetic algorithm for multi-satellite areaobservation scheduling

Aiming at the shortcomings of the traditional optimization algorithm in solving the multi-satellite regional scheduling problem such as slow convergence speed and being prone to fall into the local optimal solution, an improved adaptive genetic algorithm was proposed. The algorithm uses Monte Carlo method combined with Hamming distance to give a better initial population. According to the average Hamming distance of the population, the execution sequence of crossover and mutation operations are determined. The Sigmoid function and Gaussian function are combined to design the adaptive nonlinear crossover rate and mutation rate based on the individual fitness of the population. The dual elite retention strategy and tournament strategy are combined to ensure the inheritance of the optimal individual. Dual shutdown condition is used to improve the search efficiency of the algorithm. Finally, experiment shows that the method can significantly improve the global search ability, accelerate the convergence speed of the algorithm, and effectively improve the observation efficiency of satellites.