As work on improving cyber-physical systems (CPS) barrels ahead, Ivan Ruchkin, Ph.D., has some concerns. Loosely defined as mechanisms controlled and monitored by computer algorithms, CPS have some basic susceptibilities—rare events and unexpected circumstances can cause the system to react in unsafe or incorrect ways. Thanks to funding from the National Science Foundation—$600k over five years—Ruchkin aims to correct this. Ruchkin’s recently funded CAREER project, “Rigorous Assumption Engineering for Learning-Enabled Cyber-Physical Systems,” works toward his vision of assumption-aware CPS—ones that behave with an understanding of their own assumptions and limitations.
Part of the problem, as Ruchkin explains, is that as CPS grow in complexity and process large high-dimensional data, human engineers struggle to understand and formalize the underlying systemic assumptions, which severely limits the effectiveness of system validation and monitoring. Put another way, if human engineers aren’t sure of the assumptions powering a system, how can they ensure the system is working correctly and safely?
To make these assumptions explicit and actionable, the project will build the mathematical and algorithmic foundation for assumption awareness via specifying, validating, and responding to assumptions behind the closed-loop CPS guarantees.
To this end, this project will create an engineering methodology in three sequential thrusts: (1) discovering and representing relevant assumptions of learning-enabled CPS, (2) performing end-to-end validation of these assumptions across offline and online settings, and (3) enhancing decision-making and control to recover from online violations of these assumptions. The developed methodology will be evaluated on small-scale autonomous racing, underwater vehicles, and autonomous street traffic.
In collaboration with two centers that perform real-world CPS development at the University of Florida (UF)—the Machine Intelligence Lab and the UF Transportation Institute—the project will work towards improving our society’s ability to construct higher-performing and safer CPS for unforeseen situations. An additional outcome will be techniques, tools, and a catalog of typical assumptions that will generalize across many CPS application domains, hopefully reducing engineering costs and shortening design/deployment iterations, yielding significant economic benefits.
As is common with CAREER awards, broader societal impact is considered. Ruchkin’s project will support workforce development in the area of cyber-physical systems, training future engineers in rigorous methods using team-based lab experiences in autonomous racing and a student-led racing club. Ruchkin will also incorporate assumption specification and validation into the graduate curriculum.