NeuralFluid: Neural Fluidic System Design and Control with Differentiable Simulation
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912099109699584 |
|---|---|
| author | Li, Yifei Sun, Yuchen Ma, Pingchuan Sifakis, Eftychios Du, Tao Zhu, Bo Matusik, Wojciech |
| author_facet | Li, Yifei Sun, Yuchen Ma, Pingchuan Sifakis, Eftychios Du, Tao Zhu, Bo Matusik, Wojciech |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_14903 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | NeuralFluid: Neural Fluidic System Design and Control with Differentiable Simulation Li, Yifei Sun, Yuchen Ma, Pingchuan Sifakis, Eftychios Du, Tao Zhu, Bo Matusik, Wojciech Fluid Dynamics Artificial Intelligence Graphics 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. |
| title | NeuralFluid: Neural Fluidic System Design and Control with Differentiable Simulation |
| topic | Fluid Dynamics Artificial Intelligence Graphics |
| url | https://arxiv.org/abs/2405.14903 |