University of Glasgow
Formal verification of safety-critical systems with uncertainty for Industry 4.0 applications
Abstract
dc:description.abstractIndustry 4.0 adopts Internet of Things (IoT) and service-oriented architectures to integrate Cyber-Physical Systems (CPSs) and Enterprise Planning systems into manufacturing operations. Furthermore, manufacturing processes typically involve the composition of various modular CPSs that work as a whole, such as multiple Collaborative Robots (cobots) working together as a production line, improving the production process’s flexibility and resilience. On the other hand, it is still challenging to verify this kind of compositional process and take into account uncertainties from IoT sensors and decision-making algorithms. For example, the trustworthiness of the sensors is essential to guarantee performance, safety and product quality during operation. However, existing methodologies to test such systems often do not scale to today’s sensor networks’ complexity and dynamic nature. Formal model verification techniques are a valuable tool that allows strong reasoning about the high-level design of CPSs. However, the uncertainty exhibited by the underlying sensor networks is often ignored. Moreover, existing model-checking tools are hard to adapt to the dynamic environment of Industry 4.0 applications during the operation stage, such as an Automated Guided Vehicle (AGV) joining in accompanying the manufacturing process at run time. This thesis proposes a novel run-time formal verification framework for modular CPSs that combines sensor-level data-driven fault detection and system-level probabilistic model checking. The resulting framework can quantify sensor readings’ trustworthiness, enabling formal reasoning for system operation behaviour and reliability analysis. The proposed approach is evaluated on three use cases, including an industrial turn-mill machine equipped with a sensor network to monitor its main components continuously, a passenger lift with two sensor networks to monitor the door and cabinet car movements, and a two-cobot painting process running with Robotic Operation System (ROS). The results indicate that the proposed verification framework involving the quantified sensor’s trustworthiness enhances the accuracy of the system failure prediction and potentially optimises manufacturing processes.
Degree
thesis:*- Level dc:type.qualificationlevel
- PhD
- Grantor dc:publisher.institution
- University of Glasgow
- Year dc:date.issued
- 2025
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Xin, Xin
Subjects
dc:subject × 1Rights
- Language dc:language
- en
Identifiers
dc:identifier.*- DOI dc:identifier.doi
- https://doi.org/10.5525/gla.thesis.85121