We present an open-source library of natively differentiable physics and robotics environments, accompanied by gradient-based control methods and a benchmark-ing suite. The introduced environments allow auto-differentiation through the simulation dynamics, and thereby permit fast training of controllers. The library features several popular environments, including classical control settings from OpenAI Gym. We also provide a novel differentiable environment, based on deep neural networks, that simulates medical ventilation. We give several use-cases of new scientific results obtained using the library. This includes a medical ventilator simulator and controller, an adaptive control method for time-varying linear dynamical systems, and new gradient-based methods for control of linear dynamical systems with adversarial perturbations.
@article{arxiv.2102.09968,
title = {Deluca -- A Differentiable Control Library: Environments, Methods, and Benchmarking},
author = {Paula Gradu and John Hallman and Daniel Suo and Alex Yu and Naman Agarwal and Udaya Ghai and Karan Singh and Cyril Zhang and Anirudha Majumdar and Elad Hazan},
journal= {arXiv preprint arXiv:2102.09968},
year = {2021}
}