Neural Lyapunov Differentiable Predictive Control

About: In this work, we simultaneously learn neural control policies and neural Lyapunov functions by differentiating the constrained optimal control problem. This leads to stabilizing neural policies that respect state and action constraints.

Authors: Sayak Mukherjee, Ján Drgoňa, Aaron Tuor, Mahantesh Halappanavar, Draguna Vrabie
Venue: 61st IEEE Conference on Decision and Control

Paper: https://arxiv.org/abs/2205.10728

This work was supported by the U.S. Department of Energy through the Office of Advanced Scientific Computing Research's “Data-Driven Decision Control for Complex Systems (DnC2S)” project.
PNNL is a multi-program national laboratory operated for the U.S. Department of Energy (DOE) by Battelle Memorial Institute under Contract No. DE-AC05-76RL0-1830.