Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
This conference talk presents a novel blackbox synthesis approach for constructing continuous control functions in cyber-physical systems. Learn how researchers from Graz University of Technology and University of Oxford tackle the challenge of synthesizing controllers when analytical descriptions of the controlled plant are difficult to obtain. The presentation introduces a method that dynamically samples a limited number of test cases while guaranteeing controller correctness for a given Lipschitz bound. Discover how their algorithm employs linear interpolation models based on Delaunay triangulation to identify candidate control functions and generates additional test cases to confirm candidates or improve the model. The speakers demonstrate the effectiveness of their approach through random benchmarks and CPS examples. This 28-minute talk was presented at the VMCAI conference (January 20-21, 2025) and sponsored by ACM SIGPLAN.