Driving patterns clustering based on driving feature analysis

作者：Montazeri Gh M; Fotouhi A^*; Naderpour A

来源：Proceedings of the Institution of Mechanical Engineers - Part C: Journal of Mechanical Engineering Science , 2011, 225(C6): 1301-1317.

DOI：10.1177/2041298310392599

• driving feature
• vehicle
• fuel consumption
• exhaust emissions
• clustering

This article presents driving features analysis in order to determine superior driving features for driving conditions clustering. At first, data gathering is performed in real traffic conditions using advance vehicle location systems. Then driving data segmentation is performed and 21 driving features are defined for each driving segment. After driving feature extraction, the dependency between driving features is investigated. Influence of driving features on vehicle's fuel consumption and exhaust emissions is then studied using computer simulations. The simulation results are then verified by an experimental test. Two types of vehicles, a conventional vehicle and a hybrid electric vehicle (HEV), are simulated. Finally, the most effective driving features are determined. Two superior driving features, 'energy' and 'idle time percentage', are then used for driving segments clustering. Driving segments clustering may be utilized for driving cycle development, intelligent HEV control, etc.

• 出版日期2011