基于云计算的并行动态路径搜索算法研究

摘要

由于动态路径导航系统中总会出现预测不准确和重新计算时间长的问题,因此需要有一个高效的动态路径搜索算法作为系统的有益补充.路径搜索算法中蚁群算法具有很好的并行特性,但目前针对路径搜索中应用的蚁群算法在并行性分布方面存在重复搜索和难以找到最优解的一些缺陷.因此,本文研究针对路径搜索的更加合理的并行蚁群算法,通过合理划分数据域,使得计算结果的准确性和计算资源的利用效率都能有很大提高,最后用实验结果的对比来进一步说明算法的高效和准确性.%Facing the problem of inaccurate prediction and long computational time, the dynamic route guidance system needs an efficient dynamic path search algorithm as a useful supplement. The ant colony algorithm has good parallel characteristics, but in parallel distribution, repeated search and the increasing difficulty to find the optimal solution are serious defects. In this paper, a reasonable parallel ant colony algorithm is proposed. We improve the computing accuracy and utilization efficiency of the algorithm greatly by dividing the data domain reasonably. At the last, the effect of the algorithm is verified through experiments.