Machine Learning Forecasting of Active Nematics
arXiv:2011.00383
Abstract
Active nematics are a class of far-from-equilibrium materials characterized by local orientational order of force-generating, anisotropic constitutes. Traditional methods for predicting the dynamics of active nematics rely on hydrodynamic models, which accurately describe idealized flows and many of the steady-state properties, but do not capture certain detailed dynamics of experimental active nematics. We have developed a deep learning approach that uses a Convolutional Long-Short-Term-Memory (ConvLSTM) algorithm to automatically learn and forecast the dynamics of active nematics. We demonstrate our purely data-driven approach on experiments of 2D unconfined active nematics of extensile microtubule bundles, as well as on data from numerical simulations of active nematics.
References in corpus (20)
- Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting
- Spontaneous motion in hierarchically assembled active matter
- Hydrodynamic fluctuations and instabilities in ordered suspensions of self-propelled particles
- Topology and Dynamics of Active Nematic Vesicles
- The Geometry of Soft Materials: A Primer
- Self-organized dynamics and the transition to turbulence of confined active nematics
- Machine learning method for single trajectory characterization
- Controlling Organization and Forces in Active Matter Through Optically-Defined Boundaries
- Excitable Patterns in Active Nematics
- Large-scale chaos and fluctuations in active nematics
- Banding, Excitability and Chaos in Active Nematic Suspensions
- Correlation lengths in hydrodynamic models of active nematics
- Insensitivity of active nematic dynamics to topological constraints
- Statistical Properties of Autonomous Flows in 2D Active Nematics
- Machine-learning techniques for fast and accurate feature localization in holograms of colloidal particles
- Active nematic materials with substrate friction
- Optimal Control of Active Nematics
- Multiscale modeling and simulation of microtubule/motor protein assemblies
- Twist-induced crossover from 2D to 3D turbulence in active nematics
- Alignment of non-spherical active particles in chaotic flows