Promises of Deep Kernel Learning for Control Synthesis

Promises of Deep Kernel Learning for Control Synthesis

Reed, Robert (University of Colorado)
Laurenti, L. (TU Delft Team Luca Laurenti)
Lahijanian, Morteza (University of Colorado)

2023

Deep Kernel Learning (DKL) combines the representational power of neural networks with the uncertainty quantification of Gaussian Processes. Hence, it is potentially a promising tool to learn and control complex dynamical systems. In this letter, we develop a scalable abstraction-based framework that enables the use of DKL for control synthesis of stochastic dynamical systems against complex specifications. Specifically, we consider temporal logic specifications and create an end-to-end framework that uses DKL to learn an unknown system from data and formally abstracts the DKL model into an interval Markov decision process to perform control synthesis with correctness guarantees. Furthermore, we identify a deep architecture that enables accurate learning and efficient abstraction computation. The effectiveness of our approach is illustrated on various benchmarks, including a 5-D nonlinear stochastic system, showing how control synthesis with DKL can substantially outperform state-of-the-art competitive methods.

Artificial neural networks
Control systems
Kernel
Machine Learning
Robust Control
Scalability
Stochastic systems
Stochastic Systems
Trajectory
Uncertainty

http://resolver.tudelft.nl/uuid:7a0f3a04-48f2-4a65-8977-4df165431bf8

https://doi.org/10.1109/LCSYS.2023.3340995

2024-06-08

2475-1456

IEEE Control Systems Letters, 7, 3986-3991

Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Institutional Repository

journal article

© 2023 Robert Reed, L. Laurenti, Morteza Lahijanian