神经网络和遗传算法在动态路径诱导中的应用*

针对智能交通路径诱导目前存在的实时性差和求解效率低的问题，提出了将神经网络与遗传算法结合的动态路径诱导方法，研究了基于神经网络的交通信息实时预测方法，构造了具有时变性的路阻矩阵，解决了传统静态路阻存在时变性差等的局限性问题；探讨了基于遗传算法的最优路径求解问题，提出了适用于路径优化的编码方式、适应度函数和遗传操作算子，解决了求解效率和求解质量的平衡问题。仿真实验表明，该方法满足路径诱导的准确性、实时性和快速性要求。

Application of Neural Network and Genetic Algorithm in Dynamic Route Guidance

In order to solve the problem of bad real-time and seeking answer ineffectively,which exist in route guidance of intelligent traffic system,a dynamic route guidance method based on Neural Network(NN) and Genetic Algorithm(GA) is presented.The NN forecasting method for traffic information has been studied deeply and the time-varied road weight has composed to matrixes,which solve the limitation problem in traditional static road weight.The GA for optimal route choice is discussed,the corresponding selection,crossover and mutation operators are given.The quality and the efficient of seeking the answer have been balanced.It is indicated by simulation that the presenting method of route guidance has achieve the right, real-time and quick guidance.