Testing and verification of intelligent vehicles pose significant challenges in the field of intelligent driving. Given the complexity of current autonomous driving systems, it is necessary to expand the testing scenario library and test boundaries for intelligent vehicles. The data-driven scenario construction method enables the generation of a large number of previously unknown test scenarios and is considered the most effective approach. In response to this, this paper proposes a method for constructing a traffic flow data generation model. The method employs an improved implementation of the Auxiliary Classifier Generative Adversarial Network (ACGAN). The U-Net network serves as the generator’s network structure within the ACGAN, while the deep residual network (ResNet) serves as the discriminator’s network structure. Additionally, the model’s loss function has been enhanced. Furthermore, this paper introduces a method for evaluating the quality of the samples generated by the traffic flow generation model. The key elements of the evaluation method proposed in this paper are the consistency index and the diversity comparison method. The evaluation method confirms that the improved traffic flow generation model outperforms the original model.
KalraNPaddockSM.Driving to safety: how many miles of driving would it take to demonstrate autonomous vehicle reliability?Transp Res Part A Policy Pract2016; 94: 182–193.
2.
FengSFengYHSunHW, et al. Testing scenario library generation for connected and automated vehicles, part II: case studies. IEEE Trans Intell Transp Syst2021; 22: 5635–5647.
3.
FengSFengYHYuCH, et al. Testing scenario library generation for connected and automated vehicles, part I: methodology. IEEE Trans Intell Transp Syst2021; 22: 1573–1582.
4.
ZhuYWangJMengFQ, et al. Review on functional testing scenario library generation for connected and automated vehicles. Sensors2022; 22: 7735.
5.
DengWWLiJKRenBT, et al. A survey on automatic simulation scenario generation methods for autonomous driving. China J Highw Transp2022; 35: 316–333.
6.
QinXArechigaNDeshmukhJ, et al. Automatic testing with reusable adversarial agents. arXiv preprint arXiv:1910.13645, 2019.
7.
ZhuBZhangPXZhaoJ, et al. Review of scenario-based virtual validation methods for automated vehicles. China J Highw Transp2019; 32: 1–19.
8.
GaoFDuanJLHeYD, et al. A test scenario automatic generation strategy for intelligent driving systems. Math Probl Eng2019; 2019: 3737486.
9.
LiWPanCWZhangR, et al. AADS: augmented autonomous driving simulation using data-driven algorithms. Sci Robot2019; 4: eaaw0863.
10.
HuangXYWangPChengXJ, et al. The ApolloScape open dataset for autonomous driving and its application. IEEE Trans Pattern Anal Mach Intell2020; 42: 2702–2719.
11.
FengSSunHWYanXT, et al. Dense reinforcement learning for safety validation of autonomous vehicles. Nature2023; 615: 620.
12.
RapelliMCasettiCGagliardiG.Vehicular traffic simulation in the city of turin from raw data. IEEE Trans Mob Comput2022; 21: 4656–4666.
13.
ChaoQDengZRenJ, et al. Realistic data-driven traffic flow animation using texture synthesis. IEEE Trans Vis Comput Graph2018; 24: 1167–1178.
14.
MaXHuXWeberT, et al. Evaluation of accuracy of traffic flow generation in SUMO. Appl Sci2021; 11: 2584.
15.
YangZGAliSDingWL, et al. A way to automatically generate lane level traffic data from video in the intersections. J Adv Transp2021; 2021: 4764174.
16.
LiYDChenDW.A learning-based comprehensive evaluation model for traffic data quality in intelligent transportation systems. Multimed Tools Appl2016; 75: 11683–11698.
17.
LinPFWengJCYinBC, et al. Urban road network operation quality evaluation method based on high-frequency trajectory data. In: 21st international conference on intelligent transportation systems (ITSC), Maui, HI, USA, 4–7 November 2018, pp.3602–3607. New York: IEEE.
18.
XuYYWuXM. Traffic safety evaluation of urban expressway based on traffic flow operation characteristics. In: 6th international conference on electromechanical control technology and transportation (ICECTT 2021), Chongqing, China, 14–16 May 2021, vol. 12081, pp.920–924. SPIE.
19.
ZhangJPWangFYWangKF, et al. Data-driven intelligent transportation systems: a survey. IEEE Trans Intell Transp Syst2011; 12: 1624–1639.
20.
Medina-SalgadoBSánchez-DelaCruzEPozos-ParraP, et al. Urban traffic flow prediction techniques: a review. Sustain Comput Inf Syst2022; 35: 100739.
21.
ChenXZhangSCLiL.Multi-model ensemble for short-term traffic flow prediction under normal and abnormal conditions. IET Intell Transp Syst2019; 13: 260–268.
22.
Guerrero-IbanezJContreras-CastilloJZeadallyS.Deep learning support for intelligent transportation systems. Trans Emerg Telecommun Technol2021; 32: e4169.
23.
GalloMDe LucaG.Spatial extension of road traffic sensor data with artificial neural networks. Sensors2018; 18: 2640.
24.
SouaRKoesdwiadyAKarrayF, et al. Big-data-generated traffic flow prediction using deep learning and Dempster-Shafer theory. In: 2016 international joint conference on neural networks (IJCNN), Vancouver, BC, Canada, 24–29 July 2016, pp.3195–3202. New York: IEEE.
25.
TranNTTranVHNguyenNB, et al. On data augmentation for GAN training. IEEE Trans Image Process2021; 30: 1882–1897.
26.
FanZHuJ. Review and prospect of research on generative adversarial networks. In: 2019 IEEE 11th international conference on communication software and networks (ICCSN), Chongqing, China, 12–15 June 2019, pp.726–730. New York: IEEE.
27.
ChenYLvYWangF-Y.Traffic flow imputation using parallel data and generative adversarial networks. IEEE Trans Intell Transp Syst2020; 21: 1624–1630.
28.
LiuSQYuMJLiM, et al. The research of virtual face based on deep convolutional generative adversarial networks using TensorFlow. Physica A2019; 521: 667–680.
29.
ZhouCSZhangJSLiuJM.Lp-WGAN: using Lp-norm normalization to stabilize Wasserstein generative adversarial networks. Knowl Based Syst2018; 161: 415–424.
30.
WuXYangHChenH, et al. Long-term 4D trajectory prediction using generative adversarial networks. Transp Res Part C Emerg Technol2022; 136: 103554.
31.
WuCChenLWangGB, et al. Spatiotemporal scenario generation of traffic flow based on LSTM-GAN. IEEE Access2020; 8: 186191–186198.
32.
CoifmanBLiLZ.A critical evaluation of the Next Generation Simulation (NGSIM) vehicle trajectory dataset. Transp Res B Methodol2017; 105: 362–377.
33.
KrajewskiRBockJKloekerL, et al. The highD Dataset: a drone dataset of naturalistic vehicle trajectories on German highways for validation of highly automated driving systems. In: 2018 21st international conference on intelligent transportation systems (ITSC), Maui, HI, USA, 4–7 November 2018, pp.2118–2125. New York: IEEE.
34.
GeigerALenzPStillerC, et al. Vision meets robotics: the KITTI dataset. Int J Robot Res2013; 32: 1231–1237.
35.
GuZCLiZHDiX, et al. An LSTM-based autonomous driving model using a Waymo open dataset. Appl Sci2020; 10: 2046.
36.
WeiQLiXYSongMP.Reconstruction of irregular missing seismic data using conditional generative adversarial networks. Geophysics2021; 86: V471–V488.
