arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2412.02574cs.ROcs.AIcs.SE
收藏

生成用于测试自动驾驶系统的关键场景

Generating Critical Scenarios for Testing Automated Driving Systems

Faculty of Information Technology, University of Engineering and Technology, Vietnam National University(越南国家大学工程技术学院信息技术学院);School of Information Science, Japan Advanced Institute of Science and Technology(日本先进科学技术学院信息科学学院)

Trung-Hieu Nguyen, Truong-Giang Vuong, Hong-Nam Duong, Son Nguyen, Hieu Dinh Vo, Toshiaki Aoki, Thu-Trang Nguyen

AI总结:

提出基于强化学习的 AVASTRA,综合 ADS 内外部状态并施加启发式约束,在仿真中生成逼真关键碰撞场景,显著优于现有方法和随机搜索基线。

摘要:

自动驾驶汽车(AV)在革新交通运输方面已展现出巨大潜力,但确保其安全性和可靠性仍是一项关键挑战,尤其是在面对动态且不可预测的环境时。对自动驾驶系统(ADS)进行真实世界测试既昂贵又有风险,因此基于仿真的测试成为更受青睐的方法。本文提出 AVASTRA,这是一种基于强化学习(RL)的方法,用于在仿真环境中生成逼真的关键场景以测试 ADS。为刻画驾驶场景的复杂性,AVASTRA 通过两类信息全面表示环境:被测 ADS 的内部状态(例如 ADS 核心组件的状态、速度或加速度)以及仿真环境中周围因素的外部状态(例如天气、交通流或道路状况)。AVASTRA 训练 RL 智能体有效配置仿真环境,使 AV 处于危险情形并可能导致碰撞。我们引入一组多样化的动作,使 RL 智能体能够系统性地配置环境条件和交通参与者。此外,基于既定安全需求,我们施加启发式约束,以确保所生成测试场景的真实性和相关性。AVASTRA 在两张流行仿真地图、四种不同道路配置上进行评估。结果表明,AVASTRA 能够优于最先进方法,生成的碰撞场景多出 30% 至 115%。与基于随机搜索的基线相比,AVASTRA 的性能最高提升 275%。这些结果凸显了 AVASTRA 通过逼真、全面的关键场景生成来增强 AV 安全测试的有效性。

英文摘要:

Autonomous vehicles (AVs) have demonstrated significant potential in revolutionizing transportation, yet ensuring their safety and reliability remains a critical challenge, especially when exposed to dynamic and unpredictable environments. Real-world testing of an Autonomous Driving System (ADS) is both expensive and risky, making simulation-based testing a preferred approach. In this paper, we propose AVASTRA, a Reinforcement Learning (RL)-based approach to generate realistic critical scenarios for testing ADSs in simulation environments. To capture the complexity of driving scenarios, AVASTRA comprehensively represents the environment by both the internal states of an ADS under-test (e.g., the status of the ADS's core components, speed, or acceleration) and the external states of the surrounding factors in the simulation environment (e.g., weather, traffic flow, or road condition). AVASTRA trains the RL agent to effectively configure the simulation environment that places the AV in dangerous situations and potentially leads it to collisions. We introduce a diverse set of actions that allows the RL agent to systematically configure both environmental conditions and traffic participants. Additionally, based on established safety requirements, we enforce heuristic constraints to ensure the realism and relevance of the generated test scenarios. AVASTRA is evaluated on two popular simulation maps with four different road configurations. Our results show AVASTRA's ability to outperform the state-of-the-art approach by generating 30% to 115% more collision scenarios. Compared to the baseline based on Random Search, AVASTRA achieves up to 275% better performance. These results highlight the effectiveness of AVASTRA in enhancing the safety testing of AVs through realistic comprehensive critical scenario generation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑