Hopfield神经网络算法求解路网最优路径

为了解决经典算法在求解大规模路网最优路径时运算时间长的问题,研究了Hopfield神经网络的特点,建立了一般路网的数学模型,根据Hopfield神经网络的特点设计了适合车辆诱导的路网Hopfield神经网络最优路径算法.采用动态邻接矩阵对该算法进行了优化,减少了运算时间.把该Hopfield神经网络算法应用于所研发的车辆诱导系统的最优路径求解中,并进行了实际路网测试,结果表明应用该算法能够正确求解路网的最优路径,且比经典算法的运算效率高.

Optimal path solution of road network based on Hopfield neural network

In order to solve the problem of long time expense in computing optimal path in large road network with classical algorithms,the characteristics of Hopfield neural network were studied.A mathematic model of general road network was constructed,and a Hopfield neural network algorithm to solve optimal path in road network was designed according to the characteristics.The algorithm was optimized with dynamic adjacent matrix and reduced the calculation time.The Hopfield neural network algorithm was applied to the calculation of the optimal path in developed vehicle guidance system and tested in actual road network.Test results illuminate that the Hopfield neural network algorithm can find correct optimal path,and is more efficient than classical algorithms.