Toward an Automated Scenario-Based X-in-the-Loop Testing Framework for Connected and Automated Vehicles

被引:0
作者
Kyriakopoulos, Ioannis [1 ]
Jaworski, Pawel [1 ]
Edwards, Tim [1 ]
机构
[1] HORIBA MIRA Ltd, Nuneaton, England
来源
SAE INTERNATIONAL JOURNAL OF CONNECTED AND AUTOMATED VEHICLES | 2024年 / 5卷 / 04期
基金
“创新英国”项目;
关键词
Connected and automated vehicles; Testing and validation; Scenario-based testing; X-in-the-loop; Co-simulation; Mixed-reality;
D O I
10.4271/12-05-04-0030
中图分类号
U [交通运输];
学科分类号
08 ; 0823 ;
摘要
Emerging technologies for connected and automated vehicles (CAVs) are rapidly advancing, and there is an incremental adoption of partial automation systems in existing vehicles. Nevertheless, there are still significant barriers before fully or highly automated vehicles can enter mass production and appear on public roads. These are not only associated with the need to ensure their safe and efficient operation but also with cost and delivery time constraints. A key challenge lies in the testing and validation (T&V) requirements of CAVs, which are expected to be significantly higher than those of traditional and partially automated vehicles. Promising methodologies that can be used toward this goal are scenario-based (SBT) and X-in-the-Loop (XiL) testing. At the same time, complex techniques such as co-simulation and mixed-reality simulation could also provide significant benefits. Nevertheless, the benefits of individual solutions are likely to be significantly smaller, if considered in isolation without any supporting test automation methods. This article attempts to combine existing knowledge and state of the art to explore the development of a framework for automating the T&V needs of CAVs. To this end, the integration of the VeriCAV framework for automating SBT with the Digital CAV Proving Ground Feasibility Study (DigiCAV) XiL mixed-reality CAV development and evaluation platform has been explored. The goal of the new framework is to enable an iterative and incremental approach across all stages of CAV development through the combination of optimal scenario generation and a comprehensive XiL scenario execution environment. This article presents an overview of the new framework as well as preliminary proof of concept results.
引用
收藏
页数:12
相关论文
共 30 条
  • [1] Althoff M, 2018, IEEE INT VEH SYM, P1326, DOI 10.1109/IVS.2018.8500374
  • [2] Association for Standardization of Automation and Measuring Systems, ASAM OpenDRIVE
  • [3] Association for Standardization of Automation and Measuring Systems, ASAM OpenSCENARIO: User Guide
  • [4] CARLA Simulator-Read the Docs, CARLA 0.9.11 Documentation-OpenSCENARIO Support
  • [5] Chen R, 2018, Arxiv, DOI arXiv:1808.03089
  • [6] Cunningham M., 2015, 2015 AUSTR ROAD SAF
  • [7] Dosovitskiy A., 2017, C ROBOT LEARNING, V78, P1
  • [8] Endsley MR, 2017, J COGN ENG DECIS MAK, V11, P225, DOI 10.1177/1555343417695197
  • [9] Testing Scenario Library Generation for Connected and Automated Vehicles, Part I: Methodology
    Feng, Shuo
    Feng, Yiheng
    Yu, Chunhui
    Zhang, Yi
    Liu, Henry X.
    [J]. IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2021, 22 (03) : 1573 - 1582
  • [10] A tutorial on Bayesian nonparametric models
    Gershman, Samuel J.
    Blei, David M.
    [J]. JOURNAL OF MATHEMATICAL PSYCHOLOGY, 2012, 56 (01) : 1 - 12