Learning Safe Neural Network Controllers with Barrier Certificates
arXiv:2009.09826
Abstract
We provide a novel approach to synthesize controllers for nonlinear continuous dynamical systems with control against safety properties. The controllers are based on neural networks (NNs). To certify the safety property we utilize barrier functions, which are represented by NNs as well. We train the controller-NN and barrier-NN simultaneously, achieving a verification-in-the-loop synthesis. We provide a prototype tool nncontroller with a number of case studies. The experiment results confirm the feasibility and efficacy of our approach.
References in corpus (6)
- Safe Model-based Reinforcement Learning with Stability Guarantees
- Neural Lyapunov Control
- Learning for Safety-Critical Control with Control Barrier Functions
- Learning Control Lyapunov Functions from Counterexamples and Demonstrations
- Learning Stable Deep Dynamics Models
- Exactly Computing the Local Lipschitz Constant of ReLU Networks