CSR: Small: Dynamically Customizable Safety-Critical Embedded Systems

Washington University, Saint Louis MO

Investigators

Abstract

Major functionalities of many modern consumer products, such as smartphones are largely defined by the software that runs upon them. Such functionalities are typically customizable, and are often dynamic in the sense that they may be modified via software updates over the lifetime of the device. This project will extend the ability to perform such dynamic customization to safety-critical systems as well, thereby enabling one to customize safety-critical embedded systems and to extend their functionalities as and when a need arises for doing so. The main challenge in extending dynamic customizability to safety-critical systems arises from the need to ensure the continued correctness, both functional and temporal, of all components of a system upon updating one or more components. This project seeks to meet this need by combining formal methods, which are particularly well-suited to dealing with functional correctness, with real-time scheduling theory, which primarily deals with timing correctness. The major intellectual contribution of this project is thus integrated consideration of formal methods and scheduling theory to develop methodologies for enabling dynamic customizability in a safe and effective manner. The benefits of dynamic customizability have been made evident by its success in consumer products; the industrial applicability of a framework allowing dynamic customizability of safety-critical systems is potentially enormous, and the findings of this project will enable this. Software implementing algorithmic findings will be made available on the project web-site; all software produced for this project will be open-source. Results obtained will be incorporated into technical publications, survey papers, and perspectives articles. Teaching materials will be developed for relevant courses at the project home institution, and shared with interested instructors elsewhere. Publications, open-source software, and teaching materials produced by this research will be made available to the public on the web at URL https://sites.wustl.edu/dcsces/ and also will be retained in an Subversion (SVN) repository at Washington University in St. Louis, both for the duration of the project and for at least 5 years following its completion. Digital artifacts needed to reproduce results of this research also will be assigned Digital Object Identifiers (DOIs) and archived in Washington University's Open Scholarship web portal at https://openscholarship.wustl.edu/ as well as being provided on the https://sites.wustl.edu/dcsces/ project website along with their DOIs. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.