成熟因子映射的双系统选星方法

A Satellite Selection Method with Maturity Factor Mapping for Dual-system Satellite Navigation

传统的选星方法通常以遍历为手段,在可见星较多的情形下往往计算量很大。常规的遗传算法通常固定交叉和变异概率,产生不必要的时间消耗。针对这些问题,提出了引入成熟因子映射交叉概率和变异概率的双系统遗传选星算法,目的在于快速地找到最优解或可接受的次优解。该方法以几何精度因子(Geometric Dilution of Precision,GDOP)为适应度,构造单染色体种群,定义成熟度来指导交叉变异操作,再经过每代精英保留策略和隔代种群数量控制,最终搜索得到符合门限的可接受解。实验结果表明,在进化200代的条件下,成熟因子映射遗传算法比常规遗传算法的搜索时间平均节省约24.75%,引入种群数量控制机制后搜索时间进一步节省了约55.32%。该方法可以快速获得稳定数学期望的可用选星集合。

Traditional satellite selection method usually relies on traversal,which often results in large amounts of computation when there are many visible satellites.Conventional genetic algorithms usually fix the crossover and mutation probabilities,resulting in unnecessary time consumption.To address these problems,the authors propose a dual-system genetic satellite selection algorithm that introduces a maturity factor mapping crossover probability and mutation probability,with the aim of finding an optimal solution or an acceptable suboptimal solution quickly.The method constructs a single-chromosome population with Geometric Dilution of Precision(GDOP)as the fitness,defines the maturity level to guide the crossover mutation operation,and then searches for an acceptable solution satisfying the threshold after a per-generation elitist preservation strategy and an alternate-generation population size control.The experimental results show that,under the condition of 200 generations of evolution,the mature factor mapping genetic algorithm saves about 24.75%of the search time on average over the conventional genetic algorithm,and the introduction of the population size control mechanism further saves about 55.32%.This method can obtain the available satellite set with stable mathematical expectations quickly.