基于TsGAN的自动驾驶汽车高速公路变道切入测试场景自动生成算法  

TsGAN-Based Automatic Generation Algorithm of Lane-Change Cut-in Test Scenarios on Expressways for Autonomous Vehicles

作　　者：朱宇[1] 徐志刚[1] 赵祥模[1] 王润民[1] 曲小波[2] ZHU Yu;XU Zhigang;ZHAO Xiangmo;WANG Runmin;QU Xiaobo(IoV and CAV Testing Technology Research Institute,Chang’an University,Xi’an 710018,Shaanxi,China;School of Vehicle and Mobility,Tsinghua University,Beijing 100084,China)

机构地区：[1]长安大学车联网与智能汽车测试技术研究院,陕西西安710018 [2]清华大学车辆与运载学院,北京100084

出　　处：《华南理工大学学报（自然科学版）》2024年第8期76-88,共13页Journal of South China University of Technology(Natural Science Edition)

基　　金：陕西省杰出青年科学基金资助项目(2023-JC-JQ-45);国家重点研发计划项目(2021YFB2501200);国家自然科学基金重点资助项目(52232015);陕西省重点研发计划项目(2021LLRH-04-01-03)。

摘　　要：前方相邻车道车辆变道切入自动驾驶车辆所在车道是典型的高速公路自动驾驶高风险场景,在实际高速公路上进行该类场景的复现测试存在非常高的安全风险,虚拟仿真测试是解决该问题的最佳途径之一。为自动生成批量化的高保真高速公路变道切入测试场景,该文提出了一种基于时间序列生成对抗网络(TsGAN)的自动驾驶汽车高速公路变道切入测试场景自动生成算法。该算法以变道切入时刻两车的车头时距和侧向时距为场景危险程度评价指标,从高速公路真实轨迹数据集highD中提取4类不同危险程度的2853个变道切入场景实例,基于TsGAN构建变道切入测试场景生成模型,并利用所提取的真实轨迹数据训练模型;其后,采用所构建的模型生成前车变道轨迹和测试车变道切入时刻前的运行轨迹,通过比较所生成的变道轨迹与真实轨迹之间的分布相似性、频谱误差,验证所生成的轨迹的真实性。该文还分析了所生成的场景中变道切入时刻两车的运动关系及轨迹参数分布,检验了自动生成的场景在自然场景中的覆盖度。结果显示:从轨迹关键参数的分布来看,TsGAN模型生成的变道轨迹与真实轨迹的平均相似度达79.7%,平均频谱误差小于8%,落在最相似真实轨迹缓冲区的轨迹超过83.2%,表明所生成的轨迹具有高真实性;与所采集的真实场景相比,所生成的场景实例覆盖范围更大,在参数区间内分布更均匀,变道切入时刻前方变道车辆与测试车的车头时距和侧向时距分别降低了17.83%和16.37%(平均值),车辆轨迹参数分布区间平均增长了19.44%,表明所生成的场景在自然场景中具有较高的覆盖度;所提出的基于TsGAN的场景生成模型可模拟4种不同危险程度的变道切入测试场景,具有较强的针对性。The event wherein vehicles from the adjacent lane execute a lane-change maneuver,cutting into the lane occupied by autonomous vehicles,epitomizes a typical high-risk scenario on expressways within the domain of autonomous driving.Replicating such scenarios for testing on actual expressways involves significant safety risks.Virtual simulation test is one of the best approaches to addressing this issue.In order to automatically generate mass high-fidelity expressway lane-change cut-in test scenarios,this paper presents an automatic generation algorithm of lane-change cut-in test scenarios for autonomous vehicles based on TsGAN(Time-Series Generative Adversarial Network).In this algorithm,the time headway and lateral gap at the cut-in moment are taken as the evaluation metrics for scenario risk assessment,and 2853 instances of lane-change cut-in scenarios with four different risk levels are extracted from the real expressway trajectory dataset highD.A model for generating lanechange cut-in test scenarios is established based on TsGAN,and is trained with the extracted real trajectory data.Then,the model is employed to generate the trajectories for the lane-change vehicle and the tested vehicle before the cut-in moment.To authenticate the generated trajectories,the distribution similarity and spectral error between the generated and the real trajectories are compared.Furthermore,an in-depth analysis of the kinematic interplay between the two vehicles at the cut-in moment and the distribution of trajectory parameters in the generated scenarios is performed to validate the coverage of the generated scenarios within naturalistic settings.The findings can be summarized as follows:(1)as illustrated by the distribution of key trajectory parameters,the average similarity between the generated and the real lane-change trajectories is 79.7%,with an average spectral error less than 8%,and more than 83.2%of the generated trajectories are within the buffer of the most analogous real trajectories,indicating a notable fidelity of t

关 键 词：交通工程 自动驾驶汽车 变道切入 虚拟测试场景 生成对抗网络 

分 类 号：U463.6[机械工程—车辆工程]