arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2412.17992cs.ROcs.SYeess.SY
收藏

复杂环境中自主系统的证伪

Falsification of Autonomous Systems in Rich Environments

Rice University(莱斯大学);University of Colorado Boulder(科罗拉多大学博尔德分校)

Khen Elimelech, Morteza Lahijanian, Lydia E. Kavraki, Moshe Y. Vardi

AI总结:

针对复杂环境中自主CPS的神经网络控制器验证难题,提出将其证伪问题重构为元系统轨迹规划的元规划方法,利用RRT等采样技术并融合领域知识,在避障自动驾驶证伪实验中表现优于现有方法。

摘要:

验证依赖自动化控制器的自主信息物理系统(CPS)和人工智能(AI)智能体的行为是一个至关重要的目标。近年来,神经网络(NN)控制器展现出了巨大的潜力。遗憾的是,此类学习型控制器通常未经认证,可能导致系统出现不可预测或不安全的行为。为缓解此问题,人们在系统自动化验证方面投入了大量精力。具体而言,“黑盒测试”类工作依赖重复的系统仿真,以寻找违反规约的系统运行的证伪反例。由于运行高保真仿真计算需求大,证伪方法的目标是减少返回证伪示例所需的仿真工作量(NN推理查询)。这通常是一项巨大挑战,尤其当被测控制器训练良好时。本工作针对在不确定环境中遵循形式化规约运作的自主系统,提出了一种新颖的证伪方法。我们特别关注在语义定义的开放复杂环境中运行的CPS,此类环境会产生高维且依赖仿真的传感器观测结果。我们的方法将证伪问题重新表述为为“元系统”规划轨迹的问题,该元系统包装并封装了被测系统;我们称之为元规划方法。该问题表述可使用标准的基于采样的运动规划技术(如RRT)求解,并能逐步整合领域知识以改进搜索。我们通过一项实验研究支持了所提方法,该研究针对带有NN控制器的避障自主车进行证伪,其中元规划展现出了优于替代方法的性能。

英文摘要:

Validating the behavior of autonomous Cyber-Physical Systems (CPS) and Artificial Intelligence (AI) agents, which rely on automated controllers, is an objective of great importance. In recent years, Neural-Network (NN) controllers have been demonstrating great promise. Unfortunately, such learned controllers are often not certified and can cause the system to suffer from unpredictable or unsafe behavior. To mitigate this issue, a great effort has been dedicated to automated verification of systems. Specifically, works in the category of ``black-box testing'' rely on repeated system simulations to find a falsifying counterexample of a system run that violates a specification. As running high-fidelity simulations is computationally demanding, the goal of falsification approaches is to minimize the simulation effort (NN inference queries) needed to return a falsifying example. This often proves to be a great challenge, especially when the tested controller is well-trained. This work contributes a novel falsification approach for autonomous systems under formal specification operating in uncertain environments. We are especially interested in CPS operating in rich, semantically-defined, open environments, which yield high-dimensional, simulation-dependent sensor observations. Our approach introduces a novel reformulation of the falsification problem as the problem of planning a trajectory for a ``meta-system,'' which wraps and encapsulates the examined system; we call this approach: meta-planning. This formulation can be solved with standard sampling-based motion-planning techniques (like RRT) and can gradually integrate domain knowledge to improve the search. We support the suggested approach with an experimental study on falsification of an obstacle-avoiding autonomous car with a NN controller, where meta-planning demonstrates superior performance over alternative approaches.

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。

↑