More Like this

1. Explainable AI for Prioritizing and Deploying Defenses for Cyber-Physical System Resiliency

   https://doi.org/10.1109/TPS-ISA58951.2023.00032  

   Ray, Indrajit; Sreedharan, Sarath; Podder, Rakesh; Bashir, Shadaab Kawnain; Ray, Indrakshi (November 2023, IEEE)

   The adoption of digital technology in industrial control systems (ICS) enables improved control over operation, ease of system diagnostics and reduction in cost of maintenance of cyber physical systems (CPS). However, digital systems expose CPS to cyber-attacks. The problem is grave since these cyber-attacks can lead to cascading failures affecting safety in CPS. Unfortunately, the relationship between safety events and cyber-attacks in ICS is ill-understood and how cyber-attacks can lead to cascading failures affecting safety. Consequently, CPS operators are ill-prepared to handle cyber-attacks on their systems. In this work, we envision adopting Explainable AI to assist CPS oper-ators in analyzing how a cyber-attack can trigger safety events in CPS and then interactively determining potential approaches to mitigate those threats. We outline the design of a formal framework, which is based on the notion of transition systems, and the associated toolsets for this purpose. The transition system is represented as an AI Planning problem and adopts the causal formalism of human reasoning to asssit CPS operators in their analyses. We discuss some of the research challenges that need to be addressed to bring this vision to fruition. 

   more » « less
2. Information dynamics in the network of cyber-physical systems

   Wang, Yan (May 2020, Proceedings of 13th International Symposium on Tools and Methods of Competitive Engineering (TMCE2020))

   null (Ed.)

   Cyber-physical systems (CPS) are physical devices with highly integrated functionalities of sensing, computing, communication, and control. The levels of intelligence and functions that CPS can perform heavily rely on their intense collaboration and information sharing through networks. In this paper, the information propagation within CPS networks is studied. Information dynamics models are proposed to characterize the evolution of information processing capabilities of CPS nodes in networks. The models are based on a mesoscale probabilistic graph model, where the sensing and computing functions of CPS nodes are captured as the probabilities of correct predictions, whereas the communication functions are represented as the probabilities of mutual influences between nodes. In the proposed copula dynamics model, the information dependency among individuals is represented with joint prediction probabilities and estimated from copulas of extremal probabilities. In the proposed functional interdependency model, the correlations between prediction capabilities are captured with their functional relationships. A data-driven approach is taken to train the parameters of the information dynamics models with data from simulations. The information dynamics models are demonstrated with a simulator of CPS networks. 

   more » « less
3. Resiliency Graphs: Modelling the Interplay between Cyber Attacks and System Failures through AI Planning

   https://doi.org/10.1109/TPS-ISA62245.2024.00041  

   Bashir, Shadaab Kawnain; Podder, Rakesh; Sreedharan, Sarath; Ray, Indrakshi; Ray, Indrajit (October 2024, IEEE)

   Operation efficiency in cyber physical system (CPS) has been significantly improved by digitalization of industrial control systems (ICS). However, digitalization exposes ICS to cyber attacks. Of particular concern are cyber attacks that trigger ICS failure. To determine how cyber attacks can trigger failures and thereby improve the resiliency posture of CPS, this study presents the Resiliency Graph (RG) framework that integrates Attack Graphs (AG) and Fault Trees (FT). RG uses AI planning to establish associations between vulnerabilities and system failures thereby enabling operators to evaluate and manage system resiliency. Our deterministic approach represents both system failures and cyber attacks as a structured set of prerequisites and outcomes using a novel AI planning language. AI planning is then used to chain together the causes and the consequences. Empirical evaluations on various ICS network configurations validate the framework’s effectiveness in capturing how cyber attacks trigger failures and the framework’s scalability. 

   more » « less
4. Evaluating the effects of cyber-attacks on cyber physical systems using a hardware-in-the-loop simulation testbed

   https://doi.org/10.1109/RWEEK.2017.8088669  

   Potteiger, Bradley; Emfinger, William; Neema, Himanshu; Koutosukos, Xenofon; Tang, CheeYee; Stouffer, Keith (September 2017, 2017 Resilience Week (RWS))

   Cyber-Physical Systems (CPS) consist of embedded computers with sensing and actuation capability, and are integrated into and tightly coupled with a physical system. Because the physical and cyber components of the system are tightly coupled, cyber-security is important for ensuring the system functions properly and safely. However, the effects of a cyberattack on the whole system may be difficult to determine, analyze, and therefore detect and mitigate. This work presents a model based software development framework integrated with a hardware-in-the-loop (HIL) testbed for rapidly deploying CPS attack experiments. The framework provides the ability to emulate low level attacks and obtain platform specific performance measurements that are difficult to obtain in a traditional simulation environment. The framework improves the cybersecurity design process which can become more informed and customized to the production environment of a CPS. The developed framework is illustrated with a case study of a railway transportation system. 

   more » « less
5. Powering next-generation industry 4.0 by a self-learning and low-power neuromorphic system

   https://doi.org/doi.org/10.1145/3411295.3411302  

   Hongyu An, Dong Sam (September 2020, ACM International Conference on Nanoscale Computing and Communication)

   null (Ed.)

   With the continuous development of technologies, our society is approaching the next stage of industrialization. The Fourth Industrial Revolution also referred to as Industry 4.0, redefines the manufacturing system as a smart and connected machinery system with fully autonomous operation capability. Several advanced cutting-edge technologies, such as cyber-physical systems (CPS), the internet of things (IoT), and artificial intelligence, are believed to the essential components to realize Industry 4.0. In this paper, we focus on a comprehensive review of how artificial intelligence benefits Industry 4.0, including potential challenges and possible solutions. A panoramic introduction of neuromorphic computing is provided, which is one of the most promising and attractive research directions in artificial intelligence. Subsequently, we introduce the vista of the neuromorphic-powered Industry 4.0 system and survey a few research activities on applications of artificial neural networks for IoT. 

   more » « less