SPNets: Differentiable Fluid Dynamics for Deep Neural Networks
arXiv:1806.06094
Abstract
In this paper we introduce Smooth Particle Networks (SPNets), a framework for integrating fluid dynamics with deep networks. SPNets adds two new layers to the neural network toolbox: ConvSP and ConvSDF, which enable computing physical interactions with unordered particle sets. We use these lay- ers in combination with standard neural network layers to directly implement fluid dynamics inside a deep network, where the parameters of the network are the fluid parameters themselves (e.g., viscosity, cohesion, etc.). Because SPNets are imple- mented as a neural network, the resulting fluid dynamics are fully differentiable. We then show how this can be successfully used to learn fluid parameters from data, perform liquid control tasks, and learn policies to manipulate liquids.
arXiv commentConference on Robot Learning (CoRL) 2018
References in corpus (5)
- An Efficient Deep Learning Technique for the Navier-Stokes Equations: Application to Unsteady Wake Flow Dynamics
- Universal Planning Networks
- SPLATNet: Sparse Lattice Networks for Point Cloud Processing
- Visual Interaction Networks
- SE3-Pose-Nets: Structured Deep Dynamics Models for Visuomotor Planning and Control
Cited by in corpus (12)
- DiffCloth: Differentiable Cloth Simulation with Dry Frictional Contact
- DiffAqua: A Differentiable Computational Design Pipeline for Soft Underwater Swimmers with Shape Interpolation
- Teaching the Incompressible Navier-Stokes Equations to Fast Neural Surrogate Models in 3D
- DQC: a Python program package for Differentiable Quantum Chemistry
- Learning Pair Potentials using Differentiable Simulations
- Differentiable Molecular Simulations for Control and Learning
- Approximating the solution to wave propagation using deep neural networks
- ChainQueen: A Real-Time Differentiable Physical Simulator for Soft Robotics
- Identifying Mechanical Models through Differentiable Simulations
- Learning to Slide Unknown Objects with Differentiable Physics Simulations
- Alternating ConvLSTM: Learning Force Propagation with Alternate State Updates
- Learning to Manipulate Amorphous Materials