Formal Methods in the Design and Verification of Cyber-Physical Systems
University of Notre Dame
Description: Cyber-physical systems (CPS) are systems in which
computational
(cyber) processes interact closely with physical dynamical processes.
The design and verification of such systems requires a good understanding
of formal mathematical methods that are found in both computer science
and the traditional engineering disciplines. These formal methods
are used to model, verify, and design complex embedded systems in which
the interaction of computational and physical
processes must be approached in a holistic manner. This course introduces
first/second year graduate students to the formal methods used in
the verification and design of cyber-physical systems with specific
applications drawn from the area of networked embedded real-time control.
Topics:
- Introduction
- What is a Cyber-Physical System?
- Examples of Cyber-Physical Systems?
- Challenges
- Modeling
- Modeling Computational Processes
- Modeling Physical Processes
- Real-time Embedded Control Systems
- Requirements and Specifications
- Performance of Physical Processes (regulation)
- Safe Operation of Computational Processes (liveness,deadlock)
- Real-time logics and embedded systems
- Hybrid-system performance specifications
- Verification
- Verifying Computer Programs (fixed point theorems)
- Verification of Real-time Processes
- Verification of Hybrid Systems
- Tools
- Design
- Hierarchical Compositional Programming Methods
- Feedback Controller for Stability and Performance
- Robust Control Design
- Hierarchical and Compositional Programming
- Fault-tolerance and self-stabilizing systems
- Feedback in Computational Processes
- Example
- CSOnet System
- Process Modelling
- Controller Design
- Middleware Design
- Challenges
Grading:
Students will be expected to write critical summaries of various
papers in the open literature. Student will participate in one of several term projects
with the objective of writing a cyber-physical system paper that can be submitted
to one of the CPS-oriented conferences. 40% written summaries, 60% final presentation
Instructor:
Michael Lemmon, Dept. of Electrical Engineering,
University of Notre Dame (lemmon at nd dot edu)
Sharon Hu, Dept. of Computer Science, University of Notre Dame (hu@cse.nd.edu)
Reading List: TBD
- TBD