基于环境态势评估的智能车自主变道决策机制*

doi:10.19562/j.chinasae.qcgc.2018.09.007

摘要/Abstract

摘要： 汽车面对的是复杂高动态的行驶环境,且车载传感器信息具有不确定性,对动态环境进行正确态势评估是提高车辆,尤其是智能车行驶安全性的关键因素之一,本文中基于环境态势评估对智能车自主变道决策机制进行研究。首先基于人类驾驶认知机理对车辆环境态势评估模型进行层次化分析,然后利用动态贝叶斯网络实现态势评估,并结合最大期望效用原则实现自主变道决策,最后通过实验验证了本文方法的有效性。结果表明,该方法能在动态复杂环境和车载传感器测量数据存在偏差等信息不确定的条件下,做出正确合理的变道决策。

关键词: 智能车, 自主变道决策, 态势评估, 动态贝叶斯网络, 最大期望效用

Abstract: Motor vehicles work in the complex and highly dynamic driving environment, with the information uncertainty of onboard sensors, so the correct situation assessment of dynamic environment is one of the key factors to the safety of motor vehicles, in particular, intelligent vehicles. In this paper, the mechanism of autonomous line-change decision-making for intelligent vehicle is studied based on environment situation assessment. Firstly a hierarchy analysis on vehicle environment situation assessment model is conducted based on human driving cognitive mechanism. Then the dynamic Bayesian network is used to fulfill situation assessment, and autonomous line-change decision-making is achieved with the principle of maximum expected utility. Finally, the effectiveness of the method proposed is verified by experiments. The results show that the method proposed can make correct and reasonable decision for lane change under complex dynamic environment and the condition of information uncertainty such as measured data deviation of onboard sensors

Key words: intelligent vehicles, autonomous lane change decision, situation assessment, dynamic Bayesian network, maximum expected utility

何艳侠,尹慧琳,夏鹏飞. 基于环境态势评估的智能车自主变道决策机制^*[J]. 汽车工程, 2018, 40(9): 1048-1053.

He Yanxia, Yin Huilin & Xia Pengfei. Decision-making Mechanism of Autonomous Lane-change for
Intelligent Vehicles Based on Environment Situation Assessment[J]. , 2018, 40(9): 1048-1053.

参考文献

[1] 朱愿.基于视觉和雷达的智能车辆自主换道决策机制与控制研究[D] .北京:中国人民解放军军事医学科学院,2014.
[2] 袁伟,付锐,郭应时,等.基于视觉特性的驾驶人换道意图识别[J] .中国公路学报,2013,26(4):132-138.
[3] 彭金栓.基于视觉特性与车辆相对运动的驾驶人换道意图识别方法[D] .西安:长安大学,2012.
[4] KUMAR P, PERROLLAZ M, LEFEVRE S, et al. Learning-based approach for online lane change intention prediction[C] . Intelligent Vehicles Symposium (IV),2013 IEEE. IEEE,2013:797-802.
[5] 李玮,高德芝,段建民.智能车辆自由换道模型研究[J] .公路交通科技,2010(2):119-123.
[6] NARANJO J E, GONZALEZ C, GARCIA R de PEDRO T. Lane-change fuzzy control in autonomous vehicles for the overtaking maneuver[J] . IEEE Transactions on Intelligent Transportation Systems,2008,9(3):438-450.
[7] 袁盛玥.自动驾驶车辆城区道路环境换道行为决策方法研究[D] .北京:北京理工大学,2016.
[8] ULBRICH S, MAURER M. Towards tactical behavior planning for lane changes in automated driving[C] . Proceedings of the 18^th International IEEE Annual Conference on Intelligent Transportation Systems (ITSC 2015), Gran Canaria, Spain, Sept 2015, accepted.
[9] BRECHTEL S, GINDELE T, DILLMANN R. Probabilistic MDP-behavior planning for cars[C] . 14th Internat. IEEE Conference on Intelligent Transportation Systems (ITSC). IEEE, oct 2011:1537-1542.
[10] CHEN J. Fahrerassistenzsystem zum autonomen Spurwechsel (English title: Driver assistance system for autonomous lane changes)[D] . RWTH Aachen,2009.
[11] HABENICHT S. Development and evaluation of a maneuver based lane change assistant[D] . TU Darmstadt,2012.
[12] ULBRICH S, MAURER M. Situation assessment in tactical lane change behavior planning for automated vehicles[C] . IEEE 18th International Conference on Intelligent Transportation Systems, DOI: 10.1109/ITSC.2015.163,2015:975-981.
[13] LAUGIER C, PAROMTCHIK I E, PERROLLAZ M, et al. Probabilistic analysis of dynamic scenes and collision risks assessment to improve driving safety[J] . IEEE Intelligent Transportation Syestems Magazine,2011,3(4):4-20.
[14] 杜菲菲,冯新喜.用于态势评估的贝叶斯网络研究综述[J] .电光与控制,2010(9):42-46+59.
[15] 何友,王国宏,等.多传感器信息融合及应用[M] .北京:电子工业出版社,2000.
[16] 陈怀东.贝叶斯网络在银行操作风险管理中的应用[J] .中国经贸导刊(理论版),2017(14):31-34.
[17] 孟光磊,罗元强,梁宵,等.基于动态贝叶斯网络的空战决策方法[J] .指挥控制与仿真,2017(3):49-54.
[18] 王佳培.基于贝叶斯网络的接触网运行可靠性评估[D] .成都:西南交通大学,2017.