NeuralFluid: Neural Fluidic System Design and Control with Differentiable Simulation

TL;DR

NeuralFluid introduces a differentiable fluid simulation framework enabling efficient design and control of complex fluidic systems with dynamic boundaries, outperforming traditional methods in various benchmark tasks.

Contribution

It provides a novel differentiable Navier-Stokes solver with integrated control-shape co-design and benchmark environments for fluid system optimization.

Findings

Successful control of artificial hearts and robotic end-effectors.

Outperforms gradient-free solutions in benchmark tasks.

Enables seamless integration into learning frameworks.

Abstract

We present a novel framework to explore neural control and design of complex fluidic systems with dynamic solid boundaries. Our system features a fast differentiable Navier-Stokes solver with solid-fluid interface handling, a low-dimensional differentiable parametric geometry representation, a control-shape co-design algorithm, and gym-like simulation environments to facilitate various fluidic control design applications. Additionally, we present a benchmark of design, control, and learning tasks on high-fidelity, high-resolution dynamic fluid environments that pose challenges for existing differentiable fluid simulators. These tasks include designing the control of artificial hearts, identifying robotic end-effector shapes, and controlling a fluid gate. By seamlessly incorporating our differentiable fluid simulator into a learning framework, we demonstrate successful design, control, and learning results that surpass gradient-free solutions in these benchmark tasks.