针对遗传算法和Dijkstra算法在求解动态权值系统中最短路径时的性能问题,采用比较法,将两种算法应用在同一个实际游戏模型中,对其算法的稳定性、智能性、时间复杂度进行对比测试.游戏模型模拟了各种条件下的动态权值系统.为了使遗传算法更加可靠,通过优化其变异过程使得收敛速度更快,可靠性更高.实验数据表明,遗传算法在每张地图上的得分数以及算法所用时间普遍高于Dijkstra算法,从而得出遗传算法在求解动态权值系统中最短路径问题时稳定性和预期效果明显好于Dijkstra算法,但其时间复杂度较高的结论.%Used a comparative approach to compare the performance of the genetic algorithm with the Dijkstra algorithm when solve the shortest path problem in the dynamic weight system. Did an experiment in the actual model with these two algorithms in order to test their stability, intelligence and time complexity. The game model makes" many kinds of dynamic weight system. In order to make the genetic algorithm more reliable, the new algorithm gets a way to optimize the process of mutation to make the speed of the genetic algorithm faster and the reliability better. The experiment data shows that most data of the genetic algorithm is higher than the Dijkstra algorithm. The experiment makes a conclusion that the stability and expected result of the genetic algorithm is better than the Dijkstra algorithm in the dynamic weight system,but the time complexity of algorithm is higher than the Dijkstra algorithm.
展开▼
