论文信息 - Identification of Human Failures in the Lane Changing Process on Expressways Based on Cepstral Analysis

Identification of Human Failures in the Lane Changing Process on Expressways Based on Cepstral Analysis

Accidents can be seen as the result of an unsuccessful interplay between drivers, vehicles, traffic environments, and organizations. Human actions are an important source of vulnerability for road transportation. Current research about lane changing focuses on driving intentions. Few studies are focused on human failures during lane changing processes. In this paper, based on cepstral analysis, the characteristics of steering operation are studied to research driving failure characteristics during the lane changing processes. Then, a BP neural network model of lane changing is established to identify driving failures in lane changing processes. The identification rate of this model is 70%, which can be used to improve driving safety.

Duanfeng Chu | Hui Zhang | Chaozhong Wu | Chuan Sun | Haoran Li

[1] Kazuya Takeda,et al. Driver Modeling Based on Driving Behavior and Its Evaluation in Driver Identification , 2007, Proceedings of the IEEE.

[2] A. Pentland,et al. Towards real-time recognition of driver intentions , 1997, Proceedings of Conference on Intelligent Transportation Systems.

[3] Tomohiro Yamamura,et al. A Driver Behavior Recognition Method Based on a Driver Model Framework , 2000 .

[4] Alex Pentland,et al. Modeling and Prediction of Human Behavior , 1999, Neural Computation.

[5] H. W. Warner,et al. Manual for DREAM 3.0, Driving Reliability and Error Analysis Method. Deliverable D5.6 of the EU FP6 project SafetyNet, TREN-04-FP6TRSI2.395465/506723. , 2008 .

[6] Kazuya Takeda,et al. Driver identification using driving behavior signals , 2005, Proceedings. 2005 IEEE Intelligent Transportation Systems, 2005..

[7] Kazuya Takeda,et al. A Study of Driver Behavior Under Potential Threats in Vehicle Traffic , 2009, IEEE Transactions on Intelligent Transportation Systems.