NeuralFluid: Neural Fluidic System Design and Control with Differentiable Simulation
This work addresses the challenge of optimizing fluidic system design and control for applications such as medical devices and robotics, representing an incremental advance in differentiable simulation methods.
The authors tackled the problem of designing and controlling complex fluidic systems with dynamic solid boundaries by developing a differentiable simulation framework, achieving results that surpass gradient-free solutions in benchmark tasks like artificial heart control and robotic shape identification.
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.