RBF神经网络的行车路径代价函数建模

Radial basis function neural network modeling of the traffic path cost function

行车路线优化是城市智能交通系统的研究热点之一,对整个交通系统的优化起着重要作用.分析了影响行车时间的各种因素,结合图论中最短路径算法,建立了基于RBF神经网络的路径代价函数模型.基于该函数模型,可以计算出交通图中任意给定两地间的时间最优路径.将该模型应用于实际路况进行有效性验证,得到了有实用价值的结果,说明了该模型的正确性和有效性.

Vehicle route optimization is one of the hot topics in research on urban intelligent transportation systems （ ITS）, and it plays an important role in the optimization of the entire transportation system. This paper analyzed various lactors that attect the travel time and established a path cost tunctlon model with an radial basis lunctlon neural network, based on the shortest paths algorithms in graph theory. By this function model, the time-orientedoptimal path between any two given places on a traffic map can be calculated. The model was applied to actual traffic to validate the effectiveness, and its results are of practical value, showing the correctness and validity of themodel.