Scholarlink

<p>Cyber-physical systems (CPSs) are now widely deployed in many industrial domains, e.g., manufacturing and autonomous vehicles. To further enhance the applicability of CPSs, there comes a recent trend from both academia and industry to utilize learning-based artificial intelligence (AI) controllers for the system control process, resulting in an emerging class of AI-enabled cyber-physical systems (AI-CPSs). Although such AI-CPSs could achieve obvious performance enhancement, due to the random exploration nature and lack of systematic explanations, such AI-based techniques also bring uncertainties and safety risks to the controlled system, posing an urgent need for effective safety analysis techniques for AI-CPSs. Hence, in this work, we propose Mosaic, a model-based safety analysis framework for AI-CPSs. Mosaic first constructs a Markov decision process (MDP) model as an abstract model of the AI-CPS, which tries to characterize the behaviors of the system. Then, based on the derived model, safety analysis is designed in two aspects: online safety monitoring and offline model-guided falsification. The usefulness of Mosaic is evaluated on seven industry-level AI-CPSs, the results of which demonstrate that Mosaic is effective in providing safety monitoring to AI-CPSs and able to outperform the state-of-the-art falsification techniques, providing the basis for advanced safety analysis of AI-CPSs.</p>

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

Mosaic: model-based safety analysis for AI-enabled cyber physical system

  • 期刊论文

  • 发表日期：

    2026年2月7日

查看全文
  • Xuan Xie,
  • Jiayang Song,
  • Zhehua Zhou,
  • Fuyuan Zhang,
  • Lei Ma

摘要

Cyber-physical systems (CPSs) are now widely deployed in many industrial domains, e.g., manufacturing and autonomous vehicles. To further enhance the applicability of CPSs, there comes a recent trend from both academia and industry to utilize learning-based artificial intelligence (AI) controllers for the system control process, resulting in an emerging class of AI-enabled cyber-physical systems (AI-CPSs). Although such AI-CPSs could achieve obvious performance enhancement, due to the random exploration nature and lack of systematic explanations, such AI-based techniques also bring uncertainties and safety risks to the controlled system, posing an urgent need for effective safety analysis techniques for AI-CPSs. Hence, in this work, we propose Mosaic, a model-based safety analysis framework for AI-CPSs. Mosaic first constructs a Markov decision process (MDP) model as an abstract model of the AI-CPS, which tries to characterize the behaviors of the system. Then, based on the derived model, safety analysis is designed in two aspects: online safety monitoring and offline model-guided falsification. The usefulness of Mosaic is evaluated on seven industry-level AI-CPSs, the results of which demonstrate that Mosaic is effective in providing safety monitoring to AI-CPSs and able to outperform the state-of-the-art falsification techniques, providing the basis for advanced safety analysis of AI-CPSs.