Safe control with learned certificates: A survey of neural lyapunov, barrier, and contraction methods for robotics and control
Learning-enabled control systems have demonstrated impressive empirical performance on
challenging control problems in robotics, but this performance comes at the cost of reduced …
challenging control problems in robotics, but this performance comes at the cost of reduced …
Safe control with learned certificates: A survey of neural lyapunov, barrier, and contraction methods
Learning-enabled control systems have demonstrated impressive empirical performance on
challenging control problems in robotics, but this performance comes at the cost of reduced …
challenging control problems in robotics, but this performance comes at the cost of reduced …
Learning safe, generalizable perception-based hybrid control with certificates
C Dawson, B Lowenkamp, D Goff… - IEEE Robotics and …, 2022 - ieeexplore.ieee.org
Many robotic tasks require high-dimensional sensors such as cameras and Lidar to navigate
complex environments, but developing certifiably safe feedback controllers around these …
complex environments, but developing certifiably safe feedback controllers around these …
Safety certification for stochastic systems via neural barrier functions
FB Mathiesen, SC Calvert… - IEEE Control Systems …, 2022 - ieeexplore.ieee.org
Providing non-trivial certificates of safety for non-linear stochastic systems is an important
open problem. One promising solution to address this problem is the use of barrier functions …
open problem. One promising solution to address this problem is the use of barrier functions …
Lyapunov-stable neural-network control
Deep learning has had a far reaching impact in robotics. Specifically, deep reinforcement
learning algorithms have been highly effective in synthesizing neural-network controllers for …
learning algorithms have been highly effective in synthesizing neural-network controllers for …
Physics-informed neural network Lyapunov functions: PDE characterization, learning, and verification
We provide a systematic investigation of using physics-informed neural networks to compute
Lyapunov functions. We encode Lyapunov conditions as a partial differential equation (PDE) …
Lyapunov functions. We encode Lyapunov conditions as a partial differential equation (PDE) …
Stability analysis and controller synthesis using single-hidden-layer relu neural networks
P Samanipour, HA Poonawala - IEEE Transactions on …, 2023 - ieeexplore.ieee.org
This article presents algorithms to solve analysis and controller synthesis problems for
dynamical systems modeled as a recurrent single-hidden-layer rectified linear unit neural …
dynamical systems modeled as a recurrent single-hidden-layer rectified linear unit neural …
Learning robust output control barrier functions from safe expert demonstrations
This paper addresses learning safe output feedback control laws from partial observations of
expert demonstrations. We assume that a model of the system dynamics and a state …
expert demonstrations. We assume that a model of the system dynamics and a state …
Stability verification of neural network controllers using mixed-integer programming
In this article, we propose a framework for the stability verification of mixed-integer linear
programming (MILP) representable control policies. This framework compares a fixed …
programming (MILP) representable control policies. This framework compares a fixed …
TOOL LyZNet: A Lightweight Python Tool for Learning and Verifying Neural Lyapunov Functions and Regions of Attraction
In this paper, we describe a lightweight Python framework that provides integrated learning
and verification of neural Lyapunov functions for stability analysis. The proposed tool …
and verification of neural Lyapunov functions for stability analysis. The proposed tool …