Machine Learning-based Optimal Control for Colloidal Self-Assembly

TL;DR

This paper presents a machine learning-based optimal control framework that effectively guides colloidal self-assembly into ordered patterns, outperforming traditional methods with a 97% success rate in simulations.

Contribution

It introduces a novel combination of unsupervised learning, graph neural networks, and deep reinforcement learning for controlling colloidal self-assembly.

Findings

Superior performance over traditional order parameter methods

Achieved 97% success rate in simulation

Potential for generalization to other many-body systems

Abstract

Achieving precise control of colloidal self-assembly into specific patterns remains a longstanding challenge due to the complex process dynamics. Recently, machine learning-based state representation and reinforcement learning-based control strategies have started to accumulate popularity in the field, showing great potential in achieving an automatable and generalizable approach to producing patterned colloidal assembly. In this work, we adopted a machine learning-based optimal control framework, combining unsupervised learning and graph convolutional neural work for state observation with deep reinforcement learning-based optimal control policy calculation, to provide a data-driven control approach that can potentially be generalized to other many-body self-assembly systems. With Brownian dynamics simulations, we demonstrated its superior performance as compared to traditional order parameter-based state description, and its efficacy in obtaining ordered 2-dimensional spherical colloidal self-assembly in an electric field-mediated system with an actual success rate of 97%.