Scholarlink

This chapter examines the challenges of applying DevOps and artificial intelligence (AI) to Cyber-physical systems (CPS). While DevOps has improved time-to-market for traditional systems, scaling these practices to complex CPS with hardware-in-the-loop, software controllers, and firmware presents significant obstacles. This chapter provides a broad overview of these challenges, with subsequent chapters offering detailed solutions: (1) developing CPS-specific DevOps automation, integrating software and Hardware-in-the-Loop (HiL) testing, and using machine learning for dependability; (2) automating verification, validation, and security analysis and testing in DevOps pipelines to enhance CPS trustworthiness; (3) improving CPS adaptability with static analysis, monitoring, and self-healing; (4) creating flexible architectures and frameworks for easy deployment and compatibility in DevOps and CPS toolchains; (5) validating solutions in real-world sectors like automotive, healthcare, and railways; and (6) discussing relevant open-source technologies to encourage further development. This chapter sets the context and highlights the research gaps in DevOps for CPS, focusing on continuous integration and delivery, automated testing, AI methods, and runtime verification.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Challenges and Needs of DevOps for Cyber-Physical Systems

  • 章节

  • 发表日期：

    2026年

查看全文
  • Sebastiano Panichella,
  • Paolo Arcaini,
  • Myra B. Cohen,
  • Aitor Arrieta

摘要

This chapter examines the challenges of applying DevOps and artificial intelligence (AI) to Cyber-physical systems (CPS). While DevOps has improved time-to-market for traditional systems, scaling these practices to complex CPS with hardware-in-the-loop, software controllers, and firmware presents significant obstacles. This chapter provides a broad overview of these challenges, with subsequent chapters offering detailed solutions: (1) developing CPS-specific DevOps automation, integrating software and Hardware-in-the-Loop (HiL) testing, and using machine learning for dependability; (2) automating verification, validation, and security analysis and testing in DevOps pipelines to enhance CPS trustworthiness; (3) improving CPS adaptability with static analysis, monitoring, and self-healing; (4) creating flexible architectures and frameworks for easy deployment and compatibility in DevOps and CPS toolchains; (5) validating solutions in real-world sectors like automotive, healthcare, and railways; and (6) discussing relevant open-source technologies to encourage further development. This chapter sets the context and highlights the research gaps in DevOps for CPS, focusing on continuous integration and delivery, automated testing, AI methods, and runtime verification.