Cyber-physical systems (CPSs) combine the cyber principle of cognitive control systems with the physical elements of information perception and environmental impact, controlled or monitored with the help of computer-based algorithms. CPSs are used in an increasing number of fields, including agriculture, transport, healthcare, energy, industry, and so on. Artificial intelligence (AI) and machine learning allow intelligent CPSs to respond to various external factors by adapting their mode of operation.
Through its structure and the topics it tackles, Design automation of cyber-physical systems represents a real challenge for all those interested in developing the skills needed to use and develop modern technologies for CPS design. The case studies, as well as the diagrams, graphs, and tables, are particularly useful for illustrating the numerical data resulting from the implementation of the presented algorithms and methods, and they help clarify the analyzed concepts. The book is targeted at researchers, designers, and users alike; it covers important topics related to the design, construction, testing, and operation of modern CPSs, as well as the automation of their design being carefully systematized and tackled. It is a collaborative effort from authors recognized as valuable researchers in the analysis, design, and development of CPSs.
The first three chapters of the book are grouped in Part 1, “Design and Engineering.” Chapter 1 presents and illustrates synthesis methods and tools based on the concept of feedback function. Using platform-based design for automotive and transportation systems, chapter 2 analyzes the mapping of specifications, from high level to individual functional level, in the process of designing CPSs. Chapter 3 is dedicated to specific aspects of model-based development for networked CPSs. A standardized design specification language for networked CPSs is proposed, and its semantics described in terms of the Manna-Pnueli transition system.
Part 2, “Testing and Operation,” contains the next three chapters. In chapter 4, certain formal techniques for checking and testing CPSs, such as reachability analysis techniques and robustness-guided falsification approaches, are presented, analyzed, and thoroughly documented. Readers who are interested in testing and verifying solutions for autonomous and semi-autonomous CPSs will find a review of recent works here. Chapter 5, “Data-Driven Safety Verification of Complex Cyber-Physical Systems,” looks at road and air transport systems, energy systems, and medical systems. The relationship between data-based algorithms and numerical simulation is specified, argued, and exemplified, in the context of the calculation of the discrepancy functions of dynamic systems, the verification of hybrid CPSs, as well as the verification of models with black-box components. Testing the resilience of CPSs to potential threats is a basic requirement for assessing their vulnerability. Chapter 6 emphasizes this fact, focusing on the presentation and analysis of modeling solutions for cyber-physical resistance to attacks and the use of simulation techniques, as well as on the development and analysis of methods and utilities for the security assessment of CPSs.
The third part, “Application-Specific Design Automation Methodologies and Tools,” contains four chapters. Understanding and analyzing the vulnerabilities of large-scale CPSs, starting with individual vulnerabilities and continuing with vulnerability types, are important steps in ensuring the resilience of these systems. In chapter 7, the authors refer to the resilience of large-scale CPSs from the perspective of the optimal control of distributed systems with varying degrees of decentralization. For a distributed power supply system used for testing, the methods presented are applied, compared, and evaluated in terms of the results obtained. Model-driven engineering is one of the topics tackled in chapter 8, from the perspective of design automation for software development. At the same time, complex applications in the field of robotics are presented and analyzed; in the design phase, software solutions specific to the model-driven engineering area are used. The use of algorithms for learning engineering data is an important step in the automation of the technical design process. Supervised learning based on structural graph convolutional neural networks is presented in chapter 9, with the aim of streamlining the automation design process. A hardware/software solution with cell-level and pack-level functions for a distributed battery management system is described and analyzed in chapter 10, also from a design automation perspective.
Design automation of cyber-physical systems is excellent for researchers, academics, and practitioners who want a deeper understanding of both basic and advanced topics related to the design automation of CPSs and its vast applications. The wide range of interesting topics, combined with the depth of each chapter, makes it an essential book.