A Rapid Pattern-Recognition Method for Driving Types Using Clustering-Based Support Vector Machines

摘要

A rapid pattern-recognition approach to characterize driver'scurve-negotiating behavior is proposed. To shorten the recognition time andimprove the recognition of driving styles, a k-means clustering-based supportvector machine ( kMC-SVM) method is developed and used for classifying driversinto two types: aggressive and moderate. First, vehicle speed and throttleopening are treated as the feature parameters to reflect the driving styles.Second, to discriminate driver curve-negotiating behaviors and reduce thenumber of support vectors, the k-means clustering method is used to extract andgather the two types of driving data and shorten the recognition time. Then,based on the clustering results, a support vector machine approach is utilizedto generate the hyperplane for judging and predicting to which types the humandriver are subject. Lastly, to verify the validity of the kMC-SVM method, across-validation experiment is designed and conducted. The research resultsshow that the $ k $MC-SVM is an effective method to classify driving styleswith a short time, compared with SVM method.