A physics-informed neural network method to approximate homogeneous Lyapunov functions
Abstract
This paper applies the physics-informed neural network approach to approximate globally-defined Lyapunov functions and stabilizing controls for homogeneous dynamical systems. The advantage of this class of systems is that all analysis and design can be made locally on a suitably defined unit sphere, which corresponds perfectly to the applicability conditions of neural networks.
Origin : Files produced by the author(s)