Deluca -- A Differentiable Control Library: Environments, Methods, and Benchmarking

TL;DR

DeLuca is an open-source library offering differentiable physics environments and gradient-based control methods, enabling fast controller training and supporting scientific research in robotics and medical simulations.

Contribution

It introduces a comprehensive differentiable control library with environments, methods, and benchmarking tools, including novel medical ventilation simulation and control techniques.

Findings

Medical ventilator simulation and control demonstrated.

New gradient-based control methods for dynamical systems.

Benchmarking suite facilitates scientific research.

Abstract

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.