基于FCM的神经网络建模及其在智能驾驶中应用研究

NEURAL NETWORK MODELING BASED ON FCM CLUSTERING AND ITS APPLICATION TO INTELLIGENT DRIVING

首先对模糊 c-均值聚类算法进行了分析 ,然后把改进的 FCM算法和 RBF神经网络结合起来建模 ,得到一种映射能力较强的自组织 RBF神经网络 .最后把它应用到智能驾驶中对驾驶员的熟练程度和疲劳程度进行识别 。

The fuzzy c-means (FCM) clustering algorithm is analyzed in this paper. Then a self-organizing RBF neural network using improved FCM algorithm is proposed to increase the mapping ability. The neural network is applicable to intelligent driving, which is tested on the experimental data of the drivers' proficiency and fatigue with satisfactory results.