The project concentrates on the development of reliable physical computing systems (PCS), a class of safety-critical embedded systems that interact with the physical environment in significant ways via sensors and actuators. The target application domain is medical devices. The existing technology for PCS design does not effectively support the development of reliable and robust medical device systems as they become more complex. The new techniques will also have applicability to other domains such as avionic systems, manufacturing, and automotive controllers.
The project advances formal modeling and model-driven validation techniques to make it possible to develop highly dependable PCS. The approach is to extend hybrid systems modeling to PCS by explicitly modeling failures. These models are then used for test generation and run-time monitoring and checking. The project will also lay foundations for a rigorous certification process of such systems by developing static and dynamic dependability metrics. Finally, an open platform is being developed for creation and sharing of the artifacts of requirements, designs, models, test suites for generic infusion pumps (GIP). The platform will foster a community interested in highly dependable medical devices and ultimately enable transition of the technology into practice. A significant part of the project is the collaboration with the developers of Plug-and-Play (PnP) for operating room of the future.
In the short term, the project will improve the quality of infusion pumps and other computer-controlled medical devices, while long term impact will be a wider use of formal approaches in the engineering of embedded systems.
