Complexity Challenges in Cyber Physical Systems

Using Modeling and Simulation (M&S) to Support Intelligence, Adaptation and Autonomy

Saurabh Mittal, Andreas Tolk, Saurabh Mittal, Andreas Tolk

• Publisher: Wiley
• Year: 2020
• Print ISBN: 9781119552390
• eBook ISBN: 9781119552499
• Edition: 1
• Topic: Technology & Engineering
• Subtopic: Electrical Engineering & Telecommunications
• Index: Technology & Engineering

Offers a one-stop reference on the application of advanced modeling and simulation (M&S) in cyber physical systems (CPS) engineering

This book provides the state-of-the-art in methods and technologies that aim to elaborate on the modeling and simulation support to cyber physical systems (CPS) engineering across many sectors such as healthcare, smart grid, or smart home. It presents a compilation of simulation-based methods, technologies, and approaches that encourage the reader to incorporate simulation technologies in their CPS engineering endeavors, supporting management of complexity challenges in such endeavors.

Complexity Challenges in Cyber Physical Systems: Using Modeling and Simulation (M&S) to Support Intelligence, Adaptation and Autonomy is laid out in four sections. The first section provides an overview of complexities associated with the application of M&S to CPS Engineering. It discusses M&S in the context of autonomous systems involvement within the North Atlantic Treaty Organization (NATO). The second section provides a more detailed description of the challenges in applying modeling to the operation, risk and design of holistic CPS. The third section delves in details of simulation support to CPS engineering followed by the engineering practices to incorporate the cyber element to build resilient CPS sociotechnical systems. Finally, the fourth section presents a research agenda for handling complexity in application of M&S for CPS engineering. In addition, this text:

• Introduces a unifying framework for hierarchical co-simulations of cyber physical systems (CPS)
• Provides understanding of the cycle of macro-level behavior dynamically arising from spaciotemporal interactions between parts at the micro-level
• Describes a simulation platform for characterizing resilience of CPS

Complexity Challenges in Cyber Physical Systems has been written for researchers, practitioners, lecturers, and graduate students in computer engineering who want to learn all about M&S support to addressing complexity in CPS and its applications in today's and tomorrow's world.

Part I
Introduction

1
The Complexity in Application of Modeling and Simulation for Cyber Physical Systems Engineering

Saurabh Mittal¹ and Andreas Tolk²

¹ The MITRE Corporation, Fairborn, OH, USA

² The MITRE Corporation, Hampton, VA, USA

1.1 Introduction

Cyber Physical Systems (CPS), according to a definition provided by National Science Foundation (NSF) are hybrid networked cyber and engineered physical elements co‐designed to create adaptive and predictive systems for enhanced performance. These systems are built from, and depend upon, the seamless integration of computation and physical components. Advances in CPS are expected to enable capability, adaptability, scalability, resiliency, safety, security, and usability that will expand the horizons of these critical systems.

CPS engineering is an activity that brings these elements together in an operational scenario. Sometimes, an operational scenario may span multiple domains, for example, Smart Grid incorporating Power critical infrastructure and Water infrastructure. Intelligent home devices such as a smart washing machine utilize both the infrastructures. Another example would be Smart Transportation system wherein intelligent transportation devices interact with numerous smart vehicles to coordinate large scale traffic behaviors. Numerous such examples exist within the Internet of Things (IoT) perspective. These complex systems involve components at varying level of specifications. The constituent elements are supplied by multiple vendors and composing a solution without a formal test and evaluation infrastructure is a real challenge. Integration of functionality is not happening before the deployment, but after CPS are already deployed. CPS engineering requires a consistent model of operations that need to be supported by the compositions of various CPS contributors. CPS engineering lacks tools to design and experiment within a lab setting. How does one develop a repeatable engineering methodology to evaluate ensemble behaviors and emergent behaviors when larger systems involving critical infrastructure cannot be brought in a lab setting?

This increase in overlapping CPS capability in multitude of domains also introduces a level of complexity unprecedented in other engineered systems. The cross‐sector deployment and usage introduces risk that may have cascaded impacts in a highly networked environment. One possibility to reduce the technical risk is to remotely control the systems in the cyber environments, but the sheer number of variables and possible situations introduce complexity at multiple scales. This complexity results in test plans with limited coverage. Additional cyber physical system related issues of intelligence, adaptation, autonomy, and security make the problem even worse. The proposed solution is the enhanced use of Modeling and simulation (M&S). The M&S discipline has supported the development of complex systems since its inception. During the Spring Simulation Multi‐Conference 2017, a group of invited experts discussed general challenges in M&S of CPS. In 2018, as follow‐on panel was launched dealing with how the combination of various simulation paradigms, methods – so‐called hybrid simulation – can be utilized regarding complexity, intelligence, and adaptability of CPS.

While the focus of CPS is both on computation and physical devices, it belongs to the class of super complex systems in a man‐made world, where labels such as System of Systems (SoS), Complex Adaptive Systems (CAS), and Cyber CAS (CyCAS) are used interchangeably (Mittal 2014; Mittal and Risco‐Martín 2017a). All of them are multi‐agent systems. The constituting agents are goal‐oriented with incomplete information at any given moment and interact among themselves and with the environment. SoS is characterized by the constituent systems under independent operational and managerial control, geographical separation between the constituent systems and independent evolutionary roadmap. CAS is an SoS where constituent systems can be construed as agents that interact and adapt to the dynamic environment. Cyber CAS is a CAS that exist in a netcentric environment (for example, Internet) that incorporates human elements where distributed communication between the systems and various elements is facilitated by agreed upon standards and protocols. CPS is an SoS wherein the constituent physical and embedded systems are remotely controlled through the constituent cyber components.

Complex systems engineering identified a set of methods needed by systems engineers to govern such complex systems and cope with new challenges, like emergent properties or behavior not known in traditional systems. Many of these methods are rooted in the M&S discipline (Mittal et al. 2018). This chapter will provide an overview on the M&S methods and technologies that aid CPS engineering in the development and testing phase, and CPS governance when they are deployed in complex cyber environments. How to apply such means to enable the full potential of CPS is one of the grand challenges of our days. With this volume, we contribute to the discussion of developing a computational infrastructure for modeling, simulation, experimentation, and analytics in a transdisciplinary CPS context.

The chapter is organized as follows. Section 1.2 provides an overview on multiple modalities of CPS. Section 1.3 describes the fundamental issues with CPS engineering. Section 1.4 describes the current M&S technology, especially the co‐simulation methodology, available for CPS engineering for developing a virtual CPS environment. Section 1.5 describes the intelligence, adaptation, and autonomy aspect of CPS and how the computational element in CPS provides opportunities for advanced control and access mechanisms. Section 1.6 concludes the chapter.

1.2 Multimodal Nature of CPS

CPS are also considered as systems with integrated physical and computational capabilities that can interact with humans through variety of modalities (Baheti and Gill 2011). This ability to interact with the physical world through computational means, and by doing so expanding the capabilities of the user of the CPS, allows the CPS to interact within a team, such as enabling human‐machine‐collaborations, as well as with the environment, such as providing alternative means of locomotion – moving of the CPS –, actuation – positioning of sub‐components, such as sensors –, or manipulation – interacting with the environment.

This multimodality, the ability to interact with humans, others CPS, and the environment via a multitude of computational and physical means, is one of main sources for the complexity challenges we are coping with. It allows CPS to work in different domains and sectors, and provide their services to many different users. The same functionality can be accessed via several different interfaces to be applied in a multitude of contexts in various domains, making the validation of the CPS challenging, if not impossible. As observed in (Rajkumar et al. 2010), “…the gap between formal methods and testing needs to be bridged. Compositional verification and testing methods that explore the heterogeneous nature of CPS models are essential. V&V must also be incorporated into certification regimes” (page 735).

But validation is not the only concern. The multimodality leads to a multitude of interconnections between potentially many CPS, users, and components of the environment, creating a system of interlinked and interdependent objects. Combined with capabilities that now can be applied by CPS in the same domain, the overall complexity increases significantly.

The other side is, however, that the amount of options for an appropriate reaction in an unforeseen turn of events increases also. If many CPSs can provide ...

Table of contents

• Cover
• Table of Contents
• Preface
• Foreword
• About the Editors
• List of Contributors
• Author Biography
• Part I: Introduction
• Part II: Modeling Support to CPS Engineering
• Part III: Simulation-Based CPS Engineering
• Part IV: The Cyber Element
• Part V: Way Forward
• Cyber Physical Systems
• Index
• End User License Agreement