ARCH@ADHS最新文献

英文中文

ARCH-COMP18 Category Report: Hybrid Systems with Piecewise Constant Dynamics ARCH-COMP18分类报告:分段恒动力混合系统

ARCH@ADHS

Pub Date : 2018-09-17 DOI: 10.29007/p11g

Goran Frehse, A. Abate, D. Adzkiya, Lei Bu, Mirco Giacobbe

引用次数: 4

ARCH-COMP18 Category Report: Stochastic Modelling ARCH-COMP18类别报告:随机建模

ARCH@ADHS

Pub Date : 2018-09-17 DOI: 10.29007/7ks7

Alessandro Abate, Henk A. P. Blom, Nathalie Cauchi, S. Haesaert, A. Hartmanns, Kendra Lesser, Meeko Oishi, Vignesh Sivaramakrishnan, S. Soudjani, Cristian Ioan Vasile, Abraham P. Vinod

This report presents the results of a friendly competition for formal verification and policy synthesis of stochastic models. The friendly competition took place as part of the workshop Applied Verification for Continuous and Hybrid Systems (ARCH) in 2018. In this first edition, we present five benchmarks with different levels of complexities and stochastic favours. We make use of six different tools and frameworks (in alphabetical order): Barrier Certificates, FAUST2, FIRM-GDTL, Modest, SDCPN modelling & MC simulation and SReachTools; and attempt to solve instances of the five different benchmark problems. Through these benchmarks, we capture a snapshot on the current state-of the art tools and frameworks within the stochastic modelling domain. We also present the challenges encountered within this domain and highlight future plans which will push forward the development of more tools and methodologies for performing formal verification and optimal policy synthesis of stochastic processes.

本报告介绍了随机模型的正式验证和政策综合的友好竞争的结果。该友好竞赛是2018年连续和混合系统(ARCH)应用验证研讨会的一部分。在第一版中，我们提出了五个具有不同复杂程度和随机偏好的基准。我们使用六种不同的工具和框架(按字母顺序排列):屏障证书，FAUST2, FIRM-GDTL, Modest, SDCPN建模和MC仿真以及SReachTools;并尝试解决五个不同基准问题的实例。通过这些基准测试，我们捕捉到随机建模领域内最先进的工具和框架的当前状态的快照。我们还提出了在这一领域遇到的挑战，并强调了未来的计划，这些计划将推动更多工具和方法的发展，以执行随机过程的正式验证和最佳政策综合。

引用次数: 5

Linear Differential-Algebraic Equations (Benchmark Proposal) 线性微分代数方程(基准建议)

ARCH@ADHS

Pub Date : 2018-09-17 DOI: 10.29007/4gj7

Patrick Musau, Diego Manzanas Lopez, Hoang-Dung Tran, Taylor T. Johnson

引用次数: 2

Verification Challenges in F-16 Ground Collision Avoidance and Other Automated Maneuvers F-16地面避碰和其他自动机动的验证挑战

ARCH@ADHS

Pub Date : 2018-09-17 DOI: 10.29007/91x9

Peter Heidlauf, Alexander Collins, M. Bolender, Stanley Bak

Benchmark Proposal: The F-16 Fighting Falcon is a highly-maneuverable aircraft in production since the 1970s. Since then, several studies and books have investigated the aircraft’s performance and created simulation models. In this paper, we present some of these models as a verification challenge, providing MATLAB and Python code to simulate an F-16 performing ground collision avoidance as well as other autonomous maneuvers. The aircraft model and inner-loop controller has 16 continuous variables with piecewise nonlinear differential equations. Autonomous maneuvers are performed by an outer-loop controller using finite-state machines with guards involving the continuous variables. Passfail specifications are provided based on the aircraft flight limits and boundaries of the model. This model aims to be a starting point for analyzing detailed behaviors of aerospace systems.

基准建议:F-16战斗猎鹰是一种高机动性飞机，自20世纪70年代开始生产。从那时起，几项研究和书籍调查了飞机的性能并创建了仿真模型。在本文中，我们提出了其中一些模型作为验证挑战，提供MATLAB和Python代码来模拟F-16执行地面避碰以及其他自主机动。飞行器模型和内环控制器具有16个连续变量，具有分段非线性微分方程。自主机动由外环控制器使用有限状态机执行，该有限状态机具有涉及连续变量的保护。通行证规格是根据飞机的飞行限制和模型的边界提供的。该模型旨在成为分析航空航天系统详细行为的起点。

引用次数: 35

Lane change maneuver for autonomous vehicles (Benchmark Proposal) 自动驾驶汽车变道机动(基准方案)

ARCH@ADHS

Pub Date : 2018-09-17 DOI: 10.29007/5hxt

Nikolaos Kekatos, Daniel Hess, Goran Frehse

Lane changes are known to be risky maneuvers both for autonomous vehicles and human drivers since they require changes in longitudinal and lateral velocities in the presence of other moving vehicles. In this paper, we propose a benchmark modeling a cooperative lane change maneuver that involves four fully autonomous vehicles; three in the left lane and one in the right. The vehicle driving in the right lane aims to move to the left lane while avoiding a collision with the other vehicles. Each vehicle is equipped with sensors and can also communicate with its neighboring vehicles. The vehicle dynamics are described by a dynamic bicycle model and each vehicle is equipped with a linear low-level controller that regulates its own longitudinal and lateral behavior. To guarantee that the maneuver is safe and the traffic rules are enforced, we employ a cooperative driving control scheme (in the spirit of supervisory logic) that decides the actions of each vehicle.

众所周知，对于自动驾驶汽车和人类驾驶员来说，变道都是一种危险的操作，因为它需要在其他行驶车辆存在的情况下改变纵向和横向速度。在本文中，我们提出了一个基准模型来模拟四辆全自动驾驶汽车的协同变道机动;三个在左车道，一个在右车道。在右车道行驶的车辆的目标是在避免与其他车辆相撞的情况下驶入左车道。每辆车都配备了传感器，还可以与邻近的车辆进行通信。车辆动力学由动态自行车模型描述，每辆车都配备了一个线性低级控制器来调节其自身的纵向和横向行为。为了保证机动的安全性和交通规则的执行，我们采用了一种协作驾驶控制方案(本着监督逻辑的精神)来决定每辆车的动作。

引用次数: 3

ARCH-COMP18 Category Report: Hybrid Systems Theorem Proving ARCH-COMP18类别报告:混合系统定理证明

ARCH@ADHS

Pub Date : 2018-09-17 DOI: 10.29007/vjh3

Stefan Mitsch, A. Sogokon, Yong Kiam Tan, André Platzer, Hengjun Zhao, Xiangyu Jin, Shuling Wang, N. Zhan

引用次数: 1

ARCH-COMP18 Category Report: Bounded Model Checking of Hybrid Systems with Piecewise Constant Dynamics ARCH-COMP18类报告:分段常动力学混合系统的有界模型检验

ARCH@ADHS

Pub Date : 2018-09-17 DOI: 10.29007/q5tq

Lei Bu, Rajarshi Ray, Stefan Schupp

引用次数: 1

Discrete-Space Analysis of Partial Differential Equations 偏微分方程的离散空间分析

ARCH@ADHS

Pub Date : 2018-09-17 DOI: 10.29007/fvpp

Hoang-Dung Tran, Tianshu Bao, Taylor T. Johnson

There are numerous examples that arise and benefit from the reachability analysis problem. In cyber-physical systems (CPS), most dynamic phenomena are described as systems of ordinary differential equations (ODEs). Previous work has been done using zonotopes, support functions, and other geometric data structures to represent subsets of the reachable set and have been shown to be efficient. Meanwhile, a wide range of important control problems are more precisely modeled by partial differential equations (PDEs), even though not much attention has been paid to their reachability analyses. This reason motivates us to investigate the properties of these equations, especially from the reachability analysis and verification perspectives. In contrast to ODEs, PDEs have other space variables that also affect their behaviors and are more complex. In this paper, we study the discrete-space analysis of PDEs. Our ultimate goal is to propose a set of PDE reachability analysis benchmarks, and present preliminary analysis of different dimensional heat equations and wave equations. Finite difference methods (FDMs) are utilized to approximate the derivative at each mesh point with explicit order of errors. FDM will convert the PDE to a system of ODEs depending on the type of boundary conditions and discretization scheme chosen. After that, the problem can be treated as a common reachability problem and relevant conceptions and approaches can be applied and evaluated directly. We used SpaceEx to generate the plots and reachable regions for these equations given inputs and the series of results are shown and analyzed.

从可达性分析问题中产生并受益的例子有很多。在信息物理系统(CPS)中，大多数动态现象被描述为常微分方程(ode)系统。以前的工作已经使用分区、支持函数和其他几何数据结构来表示可达集的子集，并且已被证明是有效的。与此同时，许多重要的控制问题都是用偏微分方程(PDEs)来更精确地建模的，尽管它们的可达性分析并没有得到太多的关注。这促使我们研究这些方程的性质，特别是从可达性分析和验证的角度。与ode相比，pde具有其他空间变量，这些空间变量也会影响它们的行为，并且更加复杂。本文研究了偏微分方程的离散空间分析。我们的最终目标是提出一套PDE可达性分析基准，并对不同维度的热方程和波动方程进行初步分析。利用有限差分法(fdm)逼近每个网格点的导数，并明确误差的顺序。FDM将根据边界条件的类型和所选择的离散化方案将PDE转换为ode系统。然后，可以将该问题作为一个共同的可达性问题来对待，并直接应用和评估相关的概念和方法。我们使用SpaceEx为给定输入的这些方程生成图和可达区域，并显示和分析了一系列结果。

{"title":"Discrete-Space Analysis of Partial Differential Equations","authors":"Hoang-Dung Tran, Tianshu Bao, Taylor T. Johnson","doi":"10.29007/fvpp","DOIUrl":"https://doi.org/10.29007/fvpp","url":null,"abstract":"There are numerous examples that arise and benefit from the reachability analysis problem. In cyber-physical systems (CPS), most dynamic phenomena are described as systems of ordinary differential equations (ODEs). Previous work has been done using zonotopes, support functions, and other geometric data structures to represent subsets of the reachable set and have been shown to be efficient. Meanwhile, a wide range of important control problems are more precisely modeled by partial differential equations (PDEs), even though not much attention has been paid to their reachability analyses. This reason motivates us to investigate the properties of these equations, especially from the reachability analysis and verification perspectives. In contrast to ODEs, PDEs have other space variables that also affect their behaviors and are more complex. In this paper, we study the discrete-space analysis of PDEs. Our ultimate goal is to propose a set of PDE reachability analysis benchmarks, and present preliminary analysis of different dimensional heat equations and wave equations. Finite difference methods (FDMs) are utilized to approximate the derivative at each mesh point with explicit order of errors. FDM will convert the PDE to a system of ODEs depending on the type of boundary conditions and discretization scheme chosen. After that, the problem can be treated as a common reachability problem and relevant conceptions and approaches can be applied and evaluated directly. We used SpaceEx to generate the plots and reachable regions for these equations given inputs and the series of results are shown and analyzed.","PeriodicalId":236469,"journal":{"name":"ARCH@ADHS","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123769711","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 2

Verification of Continuous Time Recurrent Neural Networks (Benchmark Proposal) 连续时间递归神经网络的验证(基准建议)

ARCH@ADHS

Pub Date : 2018-09-17 DOI: 10.29007/6czp

Patrick Musau, Taylor T. Johnson

This manuscript presents a description and implementation of two benchmark problems for continuous-time recurrent neural network (RNN) verification. The first problem deals with the approximation of a vector field for a fixed point attractor located at the origin, whereas the second problem deals with the system identification of a forced damped pendulum. While the verification of neural networks is complicated and often impenetrable to the majority of verification techniques, continuous-time RNNs represent a class of networks that may be accessible to reachability methods for nonlinear ordinary differential equations (ODEs) derived originally in biology and neuroscience. Thus, an understanding of the behavior of a RNN may be gained by simulating the nonlinear equations from a diverse set of initial conditions and inputs, or considering reachability analysis from a set of initial conditions. The verification of continuous-time RNNs is a research area that has received little attention and if the research community can achieve meaningful results in this domain, then this class of neural networks may prove to be a superior approach in solving complex problems compared to other network architectures. Category: Academic Difficulty: High 1 Context and Origins Artificial Neural Networks have demonstrated an effective and powerful ability to achieve success in numerous contexts, such as adaptive control [43], autonomous vehicles, evolutionary robotics, pattern recognition, image classification, and nonlinear system identification and control [38] [18]. Despite this success, there have been reservations about incorporating them into safety critical systems [23] due to their susceptibility to unexpected and errant behavior from a slight perturbation in their inputs and initial conditions [42] [37]. Typically, neural networks are viewed as "black boxes" since the underlying operation of the neuron activations is often indiscernible to the creators of the network [10]. In light of these challenges, there has been significant work towards obtaining formal guarantees about the behavior of neural networks [25]. However, the majority of verification schemes have only been able to deal with neural networks that make use of piecewise-linear activation functions [7]. This is due to the great difficulty exhibited in obtaining formal guarantees for even simple properties of neural networks. In fact, neural network verification has been demonstrated to be an NP-complete problem, and while G. Frehse (ed.), ARCH18 (EPiC Series in Computing, vol. 54), pp. 196–207 Verification of Continuous Time Recurrent Neural Networks (Benchmark Proposal) Musau and Johnson techniques that make use of satisfiability modulo theories [35], mixed integer programming [41], robustness testing [4], and linear programming [13] [37] have been able to deal with small networks, they are incapable of dealing with the complexity and scale of the majority of networks present in real-life applicatio

本文介绍了连续时间递归神经网络(RNN)验证的两个基准问题的描述和实现。第一个问题处理的是原点上一个不动点吸引子的矢量场的逼近，而第二个问题处理的是一个受迫阻尼摆的系统辨识。虽然神经网络的验证是复杂的，并且大多数验证技术通常无法穿透，但连续时间rnn代表了一类网络，可以使用最初在生物学和神经科学中导出的非线性常微分方程(ode)的可达性方法进行访问。因此，可以通过模拟来自不同初始条件和输入的非线性方程，或考虑来自一组初始条件的可达性分析，来理解RNN的行为。连续时间rnn的验证是一个很少受到关注的研究领域，如果研究团体能在这一领域取得有意义的成果，那么这类神经网络可能会被证明是解决复杂问题的一种优于其他网络架构的方法。人工神经网络已经证明了在许多环境中取得成功的有效而强大的能力，例如自适应控制[43]、自动驾驶汽车、进化机器人、模式识别、图像分类和非线性系统识别与控制[38][18]。尽管取得了成功，但由于输入和初始条件的轻微扰动对意外和错误行为的敏感性，将它们纳入安全关键系统[23]一直存在保留意见[42][37]。通常，神经网络被视为“黑箱”，因为神经元激活的潜在操作通常对网络的创建者来说是无法分辨的[10]。鉴于这些挑战，人们已经在获得关于神经网络行为的正式保证方面做了大量工作[25]。然而，大多数验证方案只能处理使用分段线性激活函数的神经网络[7]。这是由于即使对神经网络的简单性质也很难得到形式保证。事实上，神经网络验证已经被证明是一个np完全问题，而G. Frehse(主编)，ARCH18 (EPiC Series In Computing, vol. 54)， pp. 196-207连续时间循环神经网络的验证(基准提案)Musau和Johnson技术利用可满足模理论[35]，混合整数规划[41]，鲁棒性测试[4]和线性规划[13][37]已经能够处理小型网络，它们无法处理现实应用中大多数网络的复杂性和规模[23]。此外，大多数验证方法只处理前馈和卷积神经网络架构。一类在验证文献中受到特别少关注的神经网络是递归神经网络。虽然前馈网络和循环网络都证明了以任何精度近似连续函数的能力[16]，但循环神经网络比前馈网络表现出一些优势[26]。通过允许在其结构中存在反馈连接，循环神经网络能够保留有关过去的信息，并使用比前馈网络更少的神经元捕获更高程度的复杂动态[5]。事实上，递归神经网络在解决事件之间存在时间关系的问题[32]方面表现出了更高的成功水平，例如捕获生物神经元的行为[28]、动态系统识别[22]、实时手势识别[3]、机器人[6,8,27,30]和语音识别[1]。因此，在这些领域，它们代表了比前馈网络更有吸引力的框架[47]。然而，由于其结构的复杂性以及其激活函数的非线性性质，目前研究文献中可用的验证方法无法应用于这些网络。因此，迫切需要能够为其运行提供正式保证的方法和先进的软件工具[23]，特别是在系统识别和安全关键系统控制的背景下。鉴于这一缺点，下面的论文提出了两个基准问题，用于验证特定类别的递归神经网络，即连续时间递归神经网络(CTRNNs)。

{"title":"Verification of Continuous Time Recurrent Neural Networks (Benchmark Proposal)","authors":"Patrick Musau, Taylor T. Johnson","doi":"10.29007/6czp","DOIUrl":"https://doi.org/10.29007/6czp","url":null,"abstract":"This manuscript presents a description and implementation of two benchmark problems for continuous-time recurrent neural network (RNN) verification. The first problem deals with the approximation of a vector field for a fixed point attractor located at the origin, whereas the second problem deals with the system identification of a forced damped pendulum. While the verification of neural networks is complicated and often impenetrable to the majority of verification techniques, continuous-time RNNs represent a class of networks that may be accessible to reachability methods for nonlinear ordinary differential equations (ODEs) derived originally in biology and neuroscience. Thus, an understanding of the behavior of a RNN may be gained by simulating the nonlinear equations from a diverse set of initial conditions and inputs, or considering reachability analysis from a set of initial conditions. The verification of continuous-time RNNs is a research area that has received little attention and if the research community can achieve meaningful results in this domain, then this class of neural networks may prove to be a superior approach in solving complex problems compared to other network architectures. Category: Academic Difficulty: High 1 Context and Origins Artificial Neural Networks have demonstrated an effective and powerful ability to achieve success in numerous contexts, such as adaptive control [43], autonomous vehicles, evolutionary robotics, pattern recognition, image classification, and nonlinear system identification and control [38] [18]. Despite this success, there have been reservations about incorporating them into safety critical systems [23] due to their susceptibility to unexpected and errant behavior from a slight perturbation in their inputs and initial conditions [42] [37]. Typically, neural networks are viewed as \"black boxes\" since the underlying operation of the neuron activations is often indiscernible to the creators of the network [10]. In light of these challenges, there has been significant work towards obtaining formal guarantees about the behavior of neural networks [25]. However, the majority of verification schemes have only been able to deal with neural networks that make use of piecewise-linear activation functions [7]. This is due to the great difficulty exhibited in obtaining formal guarantees for even simple properties of neural networks. In fact, neural network verification has been demonstrated to be an NP-complete problem, and while G. Frehse (ed.), ARCH18 (EPiC Series in Computing, vol. 54), pp. 196–207 Verification of Continuous Time Recurrent Neural Networks (Benchmark Proposal) Musau and Johnson techniques that make use of satisfiability modulo theories [35], mixed integer programming [41], robustness testing [4], and linear programming [13] [37] have been able to deal with small networks, they are incapable of dealing with the complexity and scale of the majority of networks present in real-life applicatio","PeriodicalId":236469,"journal":{"name":"ARCH@ADHS","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125940119","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

ARCH-COMP18 Category Report: Results on the Falsification Benchmarks ARCH-COMP18类别报告:伪造基准的结果

ARCH@ADHS

Pub Date : 2018-09-17 DOI: 10.29007/t85q

Adel Dokhanchi, Shakiba Yaghoubi, Bardh Hoxha, Georgios Fainekos, G. Ernst, Zhenya Zhang, Paolo Arcaini, I. Hasuo, Sean Sedwards

This report presents the outcomes of the 2018 friendly competition in the ARCH workshop for the category of falsification of temporal logic specifications over Cyber-Physical Systems.

本报告介绍了ARCH研讨会2018年关于网络物理系统时间逻辑规范伪造类别的友好竞赛的结果。

引用次数: 16

下一页尾页

类型

全部化学•材料生命科学医学物理工程技术环境•农林材料科学地球科学法学管理学化学环境科学与生态学计算机科学教育学经济学农林科学人文科学生物学数学物理与天体物理心理学综合性期刊其他工业工程理学历史学农学文学信息工程

数据库

全部 ACS Publications Elsevier ieeexplore Springer The Royal Society of Chemistry Wiley

期刊

ARCH@ADHS

全部 Acc. Chem. Res. ACS Applied Bio Materials ACS Appl. Electron. Mater. ACS Appl. Energy Mater. ACS Appl. Mater. Interfaces ACS Appl. Nano Mater. ACS Appl. Polym. Mater. ACS BIOMATER-SCI ENG ACS Catal. ACS Cent. Sci. ACS Chem. Biol. ACS Chemical Health & Safety ACS Chem. Neurosci. ACS Comb. Sci. ACS Earth Space Chem. ACS Energy Lett. ACS Infect. Dis. ACS Macro Lett. ACS Mater. Lett. ACS Med. Chem. Lett. ACS Nano ACS Omega ACS Photonics ACS Sens. ACS Sustainable Chem. Eng. ACS Synth. Biol. Anal. Chem. BIOCHEMISTRY-US Bioconjugate Chem. BIOMACROMOLECULES Chem. Res. Toxicol. Chem. Rev. Chem. Mater. CRYST GROWTH DES ENERG FUEL Environ. Sci. Technol. Environ. Sci. Technol. Lett. Eur. J. Inorg. Chem. IND ENG CHEM RES Inorg. Chem. J. Agric. Food. Chem. J. Chem. Eng. Data J. Chem. Educ. J. Chem. Inf. Model. J. Chem. Theory Comput. J. Med. Chem. J. Nat. Prod. J PROTEOME RES J. Am. Chem. Soc. LANGMUIR MACROMOLECULES Mol. Pharmaceutics Nano Lett. Org. Lett. ORG PROCESS RES DEV ORGANOMETALLICS J. Org. Chem. J. Phys. Chem. J. Phys. Chem. A J. Phys. Chem. B J. Phys. Chem. C J. Phys. Chem. Lett. Analyst Anal. Methods Biomater. Sci. Catal. Sci. Technol. Chem. Commun. Chem. Soc. Rev. CHEM EDUC RES PRACT CRYSTENGCOMM Dalton Trans. Energy Environ. Sci. ENVIRON SCI-NANO ENVIRON SCI-PROC IMP ENVIRON SCI-WAT RES Faraday Discuss. Food Funct. Green Chem. Inorg. Chem. Front. Integr. Biol. J. Anal. At. Spectrom. J. Mater. Chem. A J. Mater. Chem. B J. Mater. Chem. C Lab Chip Mater. Chem. Front. Mater. Horiz. MEDCHEMCOMM Metallomics Mol. Biosyst. Mol. Syst. Des. Eng. Nanoscale Nanoscale Horiz. Nat. Prod. Rep. New J. Chem. Org. Biomol. Chem. Org. Chem. Front. PHOTOCH PHOTOBIO SCI PCCP Polym. Chem.

﹀