Vehicle speed prediction method based on clustering and neural network

The invention relates to the technical field of intelligent traffic system vehicle speed prediction, in particular to a clustering and neural network-based vehicle speed prediction method, which comprises the following steps of: calculating Spearman coefficients among characteristics according to vehicle historical driving characteristic data including driving speed, acceleration and front vehicle distance data; determining each feature weight coefficient for clustering, determining the number of clusters according to the contour coefficient index, and clustering the historical data segments by using k-means; and further constructing a speed prediction model based on deep learning for the sample data segments in each cluster. By clustering different speed change modes and establishing a speed prediction model in a targeted manner, the precision of vehicle speed prediction in a complex driving scene is improved. 本发明涉及智能交通系统车辆速度预测技术领域，具体的说是一种基于聚类和神经网络的车辆速度预测方法，其根据车辆历史驾驶特征数据，包括行驶速度、加速度和前车距离数据，计算特征间斯皮尔曼（Spearman）