LI Shaobing， HU Xiaohui， HONG Peng
（School of Electronic and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China）
Abstract： The driving styles of the human vehicles (HVs) in the lane-changing target lane are incomplete information for the game modeling process of the autonomous vehicles (AVs), making it difficult for AVs to predict the driving decisions of HVs. A Bayesian game lane-changing prediction model is proposed, which solves incomplete information among game participants. First, vehicle acceleration’s subjective perception evaluation model is used to obtain the initial quantified value of HVs’ driving styles in the target lane, which is used for the state space’s confidence of HVs in the Bayesian game lane-changing model. Secondly, the expected safety distance based on the molecular dynamics model is used to solve the game safety benefits. Finally, the Bayesian game is transformed into a complete information static game to solve the Nash equilibrium. By comparing the matching degree of HVs’ predicted actions with actual actions, combined with the quantitative indicators of driving styles, the state space’s evaluation confidence is modified. The simulation results show that compared with the K-level game model and the robust K-level game improvement model, the Bayesian game model has a collision rate of 2% in the same traffic environment by using more stringent collision rate statistics standards, which is far lower than the 13% of the K-level game and the 7% of the robust model. The three game models account for 93%, 92% and 69% of lane-changing within 3.5 s, respectively. The proposed lane-changing model has a better effect on the decision-making risk control of AVs, and the efficiency of lane-changing within a reasonable time can also be guaranteed.
Key words： autonomous vehicle； lane-changing model； Bayesian game； driving styles； mixed traffic
References
［1］National high way traffic safety administration. National motor vehicle crash causation survey: Report to congress. National Highway Traffic Safety Administration Technical Report DOT HS, 2008, 811: 059.
［2］CAO P, XU Z, FAN Q, et al. Analysing driving efficiency of mandatory lane change decision for autonomous vehicles. IET Intelligent Transport Systems, 2019, 13(3): 506-514.
［3］WANG Z, SHI X, LI X. Review of lane-changing maneuvers of connected and automated vehicles: models, algorithms and traffic impact analyses. Journal of the Indian Institute of Science, 2019, 99(4): 589-599.
［4］RAMEZANI M, YE E. Lane density optimisation of automated vehicles for highway congestion control. Transportmetrica B: Transport Dynamics, 2019, 7(1): 1096-1116.
［5］GALCERAN E, CUNNINGHAM A G, EUSTICE R M, et al. Multipolicy decision-making for autonomous driving via changepoint-based behavior prediction: Theory and experiment. Autonomous Robots, 2017, 41(6): 1367-1382.
［6］ZIAKOPOULOS A, ROSENBLOOM T, YANNIS G. A review of the interaction between autonomous vehicles and vulnerable road users//Road Safety and Simulation International Conference (RSS), Oct. 14-17, 2019, Lowa City, Lowa, USA. 2019: 14-17.
［7］PRUEKPRASERT S, DUBUT J, ZHANG X, et al. A game theoretic approach to decision making for multiple vehicles at roundabout. 2019, arXiv preprint: arXiv: 1904.06224.
［8］GIPPSP G. A model for the structure of lane-changing decisions. Transportation Research Part B: Methodological, 1986, 20(5): 403-414.
［9］QU D Y, BING Q C, JIA Y F, et al. Molecular dynamics characteristics and models of vehicle lane changing interaction behavior. Journal of Transportation Systems Engineering and Information Technology, 2019, 19(3): 68-74.
［10］CHEN J, ZHAO P, LIANG H, et al. A multiple attribute-based decision making model for autonomous vehicle in urban environment//IEEE Intelligent Vehicles Symposium Proceedings, Jun. 8-11, 2014, MI, USA. New York: IEEE, 2014: 480-485.
［11］SHI T, WANG P, CHENG X, et al. Driving decision and control for automated lane change behavior based on deep reinforcement learning//IEEE Intelligent Transportation Systems Conference (ITSC). Oct. 27-30, 2019, Auckland, New Zealand. New York: IEEE, 2019: 2895-2900.
［12］QIAO Z, MUELLING K, DOLAN J M, et al. Automatically generated curriculum based reinforcement learning for autonomous vehicles in urban environment//IEEE Intelligent Vehicles Symposium (IV). Jun. 26-30, 2018, Changshu, Suzhou, China. New York: IEEE, 2018: 1233-1238.
［13］HOU Y, EDARA P, SUN C. Situation assessment and decision making for lane change assistance using ensemble learning methods. Expert Systems with Applications, 2015, 42(8): 3875-3882.
［14］HUBMANN C, BECKER M, ALTHOFF D, et al. Decision making for autonomous driving considering interaction and uncertain prediction of surrounding vehicles//IEEE Intelligent Vehicles Symposium (IV). Jun. 11-14, 2017, Los Angeles, CA, USA, New York: IEEE, 2017: 1671-1678.
［15］MITTELSTADT B, RUSSELL C, WACHTER S. Explaining explanations in AI//Conference on Fairness, Accountability, and Transparency. Jan. 29-31, 2019, Atlanta, GA, USA. New York: Association for Computing Machinery, 2019: 279-288.
［16］FUJIYOSHI H, HIRAKAWA T, YAMASHITA T. Deep learning-based image recognition for autonomous driving. IATSS Research, 2019, 43(4): 244-252.
［17］LI N, KOLMANOVSKY I, GIRARD A, et al. Game theoretic modeling of vehicle interactions at unsignalized intersections and application to autonomous vehicle control//Annual American Control Conference (ACC). Jun. 27-29, 2018, Milwaukee, WI, USA. New York: IEEE, 2018: 3215-3220.
［18］TIAN R, LI S, LI N, et al. Adaptive game-theoretic decision making for autonomous vehicle control at roundabouts//Conference on Decision and Control (CDC). Dec. 17-19, 2018, Fontainebleau Miami Beach, USA. New York: IEEE, 2018: 321-326.
［19］SANKAR G S, HAN K.Adaptive robust game-theoretic decision making strategy for autonomous vehicles in highway. IEEE Transactions on Vehicular Technology, 2020, 69(12): 14484-14493.
［20］KONG J, PFEIFFER M, SCHILDBACH G, et al. Kinematic and dynamic vehicle models for autonomous driving control design//Intelligent Vehicles Symposium (IV). Jun. 28-Jul. 1, 2015, Seoul, South Korea. New York: IEEE, 2015: 1094-1099.
［21］OYLER D W, YILDIZ Y, GIRARD A R, et al. A game theoretical model of traffic with multiple interacting drivers for use in autonomous vehicle development//American Control Conference (ACC). Jul.6-8, 2016, Boston, MA, USA. New York: IEEE, 2016: 1705-1710.
［22］XUE C M, TAN G Z, DING N, et al．Cooperative lane-changing model of human driving and unmanned driving based on game theory．Computer Engineering, 2017, 43(12): 261-266.
［23］STEVENS S S. Neural events and the psychophysical law. Science, 1970, 170(3962): 1043-1050.
［24］LI Y, YANG Z. Games with incomplete information and uncertain payoff: from the perspective of uncertainty theory. Soft Computing, 2019, 23(24): 13669-13678.
［25］SEDJELMACI H, SENOUCI S M, ANSARI N. Intrusion detection and ejection framework against lethal attacks in UAV-aided networks: A Bayesian game-theoretic methodology. IEEE Transactions on Intelligent Transportation Systems, 2016, 18(5): 1143-1153.
［26］KESTING A, TREIBER M, HELBING D. General lane-changing model MOBIL for car-following models. Transportation Research Record, 2007, 1999(1): 86-94.
［27］YANG D, ZHENG S, WEN C, et al. A dynamic lane-changing trajectory planning model for automated vehicles. Transportation Research Part C: Emerging Technologies, 2018, 95: 228-247.
［28］HARTWICH F, BEGGIATO M, KREMS J F. Driving comfort, enjoyment and acceptance of automated driving-effects of drivers’ age and driving style familiarity. Ergonomics, 2018, 61(8): 1017-1032.
［29］ASKARI G, GORDJI M E, PARK C. The behavioral model and game theory. Palgrave Communications, 2019, 5(1): 57-65.
［30］HARSANYIJ C. Games with incomplete information played by “Bayesian” players, I-II: Part I. The basic model. Management Science, 2004, 50: 1804-1817.
［31］HE W, SUN Y. Pure-strategy equilibria in Bayesian games. Journal of Economic Theory, 2019, 180: 11-49.
基于贝叶斯博弈的自动驾驶车辆换道预测模型
李绍兵， 胡晓辉， 洪鹏
（兰州交通大学 电子与信息工程学院， 甘肃 兰州 730070）
摘要：在混合交通环境下， 人类驾驶员换道目标车道的驾驶风格， 是自动驾驶车辆博弈建模过程中的不完全信息， 导致自动驾驶车辆无法预测人类驾驶员的驾驶决策。 本文提出了一种解决博弈参与者间不完全信息的贝叶斯博弈换道预测模型。 首先， 使用车辆加速度主观感受评估模型， 得到目标车道人类驾驶员驾驶风格的初始量化值， 并用于贝叶斯博弈换道模型中人类驾驶员状态空间的置信度。 其次， 使用基于分子动力学模型的期望安全距离求解博弈安全收益。 最后， 将贝叶斯博弈转化为完全信息静态博弈求解纳什均衡， 通过对比人类驾驶员预测动作和实际动作的匹配度， 结合驾驶风格量化指标修改其状态空间评估置信度。 仿真结果表明： 与K级博弈模型和具有鲁棒性的K级博弈改进模型相比， 在相同的交通环境中， 采用更严格的碰撞率统计标准， 贝叶斯博弈模型的碰撞率为2%， 远低于K级博弈的13%和鲁棒模型的7%， 三种博弈模型在3.5秒内完成换道的占比分别为93%， 92%和69%。 本文所提换道模型在自动驾驶车辆决策风险控制上具有更好的效果， 在合理时间范围内的换道效率也可以得到保障。
关键词：自动驾驶； 换道模型； 贝叶斯博弈； 驾驶风格； 混合交通
引用格式：LI Shaobing， HU Xiaohui， HONG Peng． Lane-changing prediction model of autonomous vehicle based on Bayesian game． Journal of Measurement Science and Instrumentation， 2023， 14（2）： 242-252. DOI： 10．3969／j．issn．1674-8042．2023．02．014
[full text view]
