Driver lane change decision analysis and intention recognition algorithm
DING Jieyun1, DANG Ruina2, WANG Jianqiang1, LI Keqiang1
1. State Key Laboratory of Automotive Safety and Energy, Tsinghua University, Beijing 100084, China;
2. China North Vehicle Research Institute, Beijing 100072, China
Abstract：An accurate lane change intention recognition algorithm is developed to improve real-time performance. The algorithm analyzes the drivers' lane change decisions to develop a new symbol and a comprehensive decision index (CDI) based on fuzzy theory to assess the probability that the driver will change lanes. Then, the driver intention recognition algorithm is designed based on a hidden Markov model. Using the new symbol as well as representative lateral motion parameters as observed signals, and the driver's intention as the hidden state, a hidden Markov model is built and trained. The driver's intention is recognized by the HMM decoding method. Lane change data collected on a driving simulator are used to verify the overall algorithm performance. The results show that the algorithm with the CDI as one of the observation signals both guarantees the accuracy of the recognition results and improves the real-time performance.
[1] 李克强. 汽车技术的发展动向及我国的对策[J]. 汽车工程, 2009, 31 (11): 1005-1016. LI Keqiang. Automotive technology development and countermeasures of China [J]. Automotive Engineering, 2009, 31 (11): 1005-1016. (in Chinese)
[2] Amditis A, Bimpas M, Thomaidis G, et al. A situation-adaptive lane-keeping support system: Overview of the SAFELANE approach [J]. IEEE Transactions on Intelligent Transportation Systems, 2010, 11 (3): 617-629.
[3] Kuge N, Yamamura T, Shimoyama O, et al. A Driver Behavior Recognition Method Based on a Driver Model Framework [R]. SAE 2000-01-0349. Detroit, MI, USA: SAE International, 2000.
[4] Mizushima T, Raksincharoensak P, Nagai M. Direct yaw-moment control adapted to driver behavior recognition [C]//Proc of the SICE-ICASE International Joint Conference. Busan, Korea, 2006: 534-539.
[5] McCall J C, Wipf D P, Trivedi M M, et al. Lane change intent analysis using robust operators and sparse Bayesian learning [J]. IEEE Transactions on Intelligent Transportation Systems, 2007, 8 (3): 431-440.
[6] LIANG Yulan, Reyes M L, Lee J D. Real-time detection of driver cognitive distraction using support vector machines [J]. IEEE Transactions on Intelligent Transportation Systems, 2007, 8 (2): 340-350.
[7] 党睿娜. 具有换道辅助功能的车辆自适应巡航控制[D]. 北京: 清华大学, 2013.DANG Ruina. Vehicular Adaptive Cruise Control with Lane Change Assist [D]. Beijing: Tsinghua University, 2013. (in Chinese)
[8] 党睿娜, 王建强, 李克强, 等. 高速公路行驶条件下的驾驶人换道特性[J]. 清华大学学报: 自然科学版, 2013, 53 (10): 1481-1485.DANG Ruina, WANG Jianqiang, LI Keqiang, et al. Driver lane change characteristics for various highway driving conditions [J]. Journal of Tsinghua University: Science and Technology, 2013, 53 (10): 1481-1485. (in Chinese)
[9] Kondoh T, Yamamura T, Kitazaki S, et al. Identification of visual cues and quantification of drivers' perception of proximity risk to the lead vehicle in car-following situations [J]. Journal of Mechanical Systems for Transportation and Logistics, 2008, 1 (2): 170-180.
[10] Shaout A, Colella D, Awad S. Advanced driver assistance systems: Past, present and future [C]//Proc of the 7th International Computer Engineering Conference. Giza, Egypt, 2011: 72-82.
[11] DANG Ruina, DING Jieyun, BO Su, et al. A lane change warning system based on V2V communication [C]//Proc of the 17th International Intelligent Transportation Systems Conference. Qingdao, China, 2014: 1923-1928.
[12] DING Jieyun, DANG Ruina, WANG Jianqiang, et al. Driver intention recognition method based on comprehensive lane-change environment assessment [C]//Proc of 2014 IEEE Intelligent Vehicle Symposium. Dearborn, MI, USA, 2014: 214-220.
2015, Vol. 55 
