PhysCov: Physical Test Coverage for Autonomous Vehicles

PhysCov: Physical Test Coverage for Autonomous Vehicles
复制标题

PhysCov:自动驾驶汽车的物理测试覆盖率

DOI:
10.1145/3597926.3598069
复制
发表时间:
2023
期刊:
Proceedings of the 32nd ACM SIGSOFT International Symposium on Software Testing and Analysis
影响因子:
--
通讯作者:
Elbaum, Sebastian
Elbaum, Sebastian
中科院分区:
--
文献类型:
--
作者:
Hildebrandt, Carl;von Stein, Meriel;Elbaum, Sebastian

文献摘要

参考文献

被引文献

相似文献

充分练习自动驾驶汽车的行为是其验证的基础。然而,量化自动驾驶汽车的测试充分性具有挑战性,因为系统的行为受到其状态和物理环境的影响。为了应对这一挑战,我们的工作建立在两个见解的基础上。首先,自动驾驶车辆感测的数据提供了物理环境输入的独特空间特征。其次,考虑到车辆的当前状态,位于自动驾驶车辆物理可达区域之外的输入与其行为的相关性较小。基于这些见解,我们引入了一种抽象概念,可以计算物理环境状态覆盖指标 PhysCov。该抽象将传感器读数与基于车辆状态和动态的物理可达性分析相结合,以确定可能影响自动驾驶车辆的环境区域。然后,它通过可参数化的几何近似来表征该区域,该几何近似可以以质量换取成本。具有相同特征的测试被认为具有相似的内部状态并暴露于相似的环境,因此可能会执行相同的行为集,而具有不同特征的测试将增加 PhysCov。对两个模拟系统和一个真实系统数据集的研究检验了 PhysCovs 量化自动驾驶汽车测试套件的能力,展示其表征成本和精度,调查其与发现的故障的相关性和测试选择的潜力,并评估其区分现实场景的能力。
Adequately exercising the behaviors of autonomous vehicles is fundamental to their validation. However, quantifying an autonomous vehicle’s testing adequacy is challenging as the system’s behavior is influenced both by itsstateas well as itsphysical environment. To address this challenge, our work builds on two insights. First, data sensed by an autonomous vehicle provides a unique spatial signature of the physical environment inputs. Second, given the vehicle’s current state, inputs residing outside the autonomous vehicle’s physically reachable regions are less relevant to its behavior. Building on those insights, we introduce an abstraction that enables the computation of a physical environment-state coverage metric,PhysCov. The abstraction combines the sensor readings with a physical reachability analysis based on the vehicle’s state and dynamics to determine the region of the environment that may affect the autonomous vehicle. It then characterizes that region through a parameterizable geometric approximation that can trade quality for cost. Tests with the same characterizations are deemed to have had similar internal states and exposed to similar environments and thus likely to exercise the same set of behaviors, while tests with distinct characterizations will increasePhysCov. A study on two simulated and one real system’s dataset examinesPhysCovs’s ability to quantify an autonomous vehicle’s test suite, showcases its characterization cost and precision, investigates its correlation with failures found and potential for test selection, and assesses its ability to distinguish among real-world scenarios.
实现自动驾驶汽车覆盖范围保证的系统级测试
DOI: 10.1109/models.2019.00-12
发表时间: 2019
期刊: 2019 ACM/IEEE 22nd International Conference on Model Driven Engineering Languages and Systems (MODELS)
影响因子: --
作者:
I. Majzik;Oszkár Semeráth;C. Hajdu;Kristóf Marussy;Z. Szatmári;Zoltán Micskei;András Vörös;Aren A. Babikian;Dániel Varró
通讯作者: Dániel Varró
使用傅立叶变换对不受控线性系统进行正向随机可达性分析
DOI: 10.1145/3049797.3049818
发表时间: 2016
期刊: Proceedings of the 20th International Conference on Hybrid Systems: Computation and Control
影响因子: --
作者:
Abraham P. Vinod;B. Homchaudhuri;Meeko Oishi
通讯作者: Meeko Oishi
用于自动驾驶车辆回归测试的自动场景生成
DOI: --
发表时间: 2017
期刊: 2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC)
影响因子: --
作者:
Elias Rocklage;Heiko Kraft;Abdullah Karatas;J. Seewig
通讯作者: J. Seewig
基于程序的软件测试数据充分性标准的评估
DOI: 10.1145/62959.62963
发表时间: 1988
期刊: Commun. ACM
影响因子: --
作者:
E. Weyuker
通讯作者: E. Weyuker
设置可达性分析的传播技术
DOI: --
发表时间: 2021
期刊: Annu. Rev. Control. Robotics Auton. Syst.
影响因子: --
作者:
M. Althoff;Goran Frehse;A. Girard
通讯作者: A. Girard