中文摘要：

针对具有侧摆能力的对地观测卫星的自主任务调度问题,对卫星自主任务调度问题和约束条件进行了描述,针对卫星自主任务调度NP-hard的特点,构建了基于目标收益及多约束卫星任务调度模型。设计了一种改进的遗传算法,从遗传操作的各个部分进行算法优化。首先将小区间法应用于初始种群生成,保证了种群的多样性,并且交叉和变异算子均引入自适应概率;同时采用两代竞争技术来避免“早熟”现象,提高算法的效率和鲁棒性。算法还采用最优保留策略用来保存进化中的最优解,使得算法收敛于全局最优。对局部多冲突观测任务应用该改进遗传算法,并针对区域密集目标的观测问题设计了仿真试验,与传统模拟退火算法及免疫蚁群遗传混合算法进行了比较,验证了该算法的有效性和收敛效果。

英文摘要：

Aiming at solving the problem of autonomous task scheduling of earth observing satellites with the ability of swinging, satellite autonomous task scheduling problem and constraints were described. A single-objective multi-constraints model was built according to the NP-hard character of satellite autonomous task scheduling problem. An adapted genetic algorithm was designed. All of the genetic operations of genetic algorithms were optimized. Firstly, mini-region method was applied to generate of the original population to ensure the diversity of population. Adaptive probabilities were used for crossover and mutation operation. Two generations competitive technology was used to avoid the premature and improve the efficiency and the robustness of the algorithm. The algorithm also uses the optimization reserved strategy to preserve the optimal solution, which makes the algorithm converge to the global. The adapted genetic algorithm was applied to the local multi-conflict tasks observation and designed simulation experiments of the observation of regional dense targets. The results are compared with results of simulated annealing algorithm and immune ant colony genetic algorithm, and it shows that the proposed algorithm is more effective and it has a better convergence.