Searcharxiv⌕ Search

arXiv subjects

Beena

Publications and source records attributed to Beena.

2 recordsLinked to original sources

Studying, Identifying, and Fixing Hidden Technical Debt in AI-Intensive Cyber-Physical Systems

Artificial Intelligence (AI) components are increasingly pervasive in several software systems, including Cyber-Physical Systems (CPSs). AI-CPS are used in several domains, including autonomous vehicles, industry, home automation, robotics, and healthcare. Being composed of hardware, AI components, and conventional modules, AI-CPS can exhibit technical debt (TD) that is peculiar and potentially more challenging than that of conventional systems. This thesis aims to characterize AI-CPS TD and propose approaches for its identification and repair. In a first phase, we characterize AI-CPS TD by analyzing AI ecosystems and AI-CPS repositories, as well as interviewing developers. Based on the acquired knowledge, we define approaches to identify and mitigate such TD. Finally, we plan to develop and validate an automated tool that supports agentic AI solutions to monitor, govern, and repay AI-CPS TD.

cs.SE↗

Uncertainty in Cyber-Physical Systems: A Literature Survey

Cyber-Physical Systems (CPSs) operate in dynamic, dependable environments where sensing, communication, decision-making, control, and assurance are affected by uncertainty. This paper characterizes CPS uncertainty through a literature survey comprising 186 primary studies. We (i) characterize uncertainty types, (ii) identify uncertainty sources, (iii) relate uncertainty to life-cycle phases in which they occur, and (iv) overview the possible uncertain mitigation strategies. The findings show that uncertainty is most often conceptualized as an assurance issue. Safety and specification uncertainty is the most prevalent category, followed by environmental and model/epistemic uncertainty. Domain mapping reveals broad cross-domain engagement, while safety-critical sectors such as autonomous systems, energy systems, and robotics CPSs display domain-specific uncertainty patterns. The life-cycle analysis indicates that uncertainty is not limited to the development or testing phases but also arises during runtime operation, underscoring the need for continuous assurance approaches. Mitigation strategies primarily include runtime monitoring, robust and adaptive control, and probabilistic reasoning. In summary, this survey contributes to a thorough characterization of CPS uncertainty and provides a discussion of articles on this topic.

cs.SE↗