高速公路自动驾驶渗透率对事件响应仿真与参数优化  

Simulation and Parameter Optimization of Highway Autonomous Driving Penetration on Event Response

作　　者：张鲁宁 王景升 ZHANG Luning;WANG Jingsheng(School of Traffic Management People's Public Security University of China,Beijing 100038,China)

机构地区：[1]中国人民公安大学交通管理学院,北京100038

出　　处：《交通工程》2025年第4期1-11,共11页Journal of Transportation Engineering

摘　　要：随着自动驾驶技术的发展,自动驾驶车辆(CAV)与人工驾驶车辆(HV)在混合交通流中的交互日益复杂,尤其是在高速公路异常事件中,其对交通流的影响亟待深入研究。为此,本研究采用Carl a-SUMO联合仿真平台,构建环形高速公路场景,模拟真实高速公路行驶特性,并设置事故区域以模拟异常事件。通过与真实交通数据的对比,利用遗传算法对仿真模型的关键参数进行优化,以提高仿真的准确性和可靠性。研究结果表明,随着CAV渗透率的增加,交通流的流畅性和稳定性显著提升,但系统在突发事件中的适应性可能下降。敏感性分析显示,策略性换道(lcStrategic)和减速度(decel)是影响交通系统性能的主要因素。具体而言,当CAV渗透率从0增加到100%时,最大排队长度显著缩短,拥堵持续时间逐步减少,而空间平均速度显著上升。这些发现为CAV与HV协同环境下的交通系统优化提供科学依据,表明通过合理优化CAV参数,可改善混合交通流的稳定性和安全性。With the development of automated driving technology,the interaction between a self-driving vehicle(CAV)and a human-driven vehicle(HV)in a mixed traffic flow has become increasingly complex,especially in the case of highway anomalous events,whose impact on the traffic flow urgently needs in-depth study.To this end,this study adopts the Carl a-SUMO joint simulation platform to construct a circular freeway scenario to simulate the driving characteristics of real freeways,and sets up an accident area to simulate abnormal events.By comparing with real traffic data,the key parameters of the simulation model are optimized using genetic algorithm to improve the accuracy and reliability of the simulation.The results show that with the increase of CAV penetration,the smoothness and stability of traffic flow are significantly improved,but the adaptability of the system in unexpected events may decrease.Sensitivity analysis shows that strategic lane changing(lcStrategic)and deceleration(decel)are the main factors affecting the performance of the traffic system.Specifically,as CAV penetration increases from 0 to 100%,maximum queue length is significantly reduced,congestion duration is progressively reduced,and spatially averaged speed significantly increases.These findings provide a scientific basis for the optimization of the traffic system under the cooperative environment of CAV and HV,and show that the stability and safety of mixed traffic flow can be effectively improved by rationally optimizing CAV parameters.

关 键 词：Carl a-SUMO 参数优化 帕累托前沿 贝叶斯网络 混合交通流 

分 类 号：U491.14[交通运输工程—交通运输规划与管理]