Reinforcement Learning of Artificial Microswimmers

Abstract

The behavior of living systems is based on the experience they gained through their interactions with the environment [1]. This experience is stored in the complex biochemical networks of cells and organisms to provide a relationship between a sensed situation and what to do in this situation [2-4]. An implementation of such processes in artificial systems has been achieved through different machine learning algorithms [5, 6]. However, for microscopic systems such as artificial microswimmers which mimic propulsion as one of the basic functionalities of living systems [7, 8] such adaptive behavior and learning processes have not been implemented so far. Here we introduce machine learning algorithms to the motion of artificial microswimmers with a hybrid approach. We employ self-thermophoretic artificial microswimmers in a real world environment [9, 10] which are controlled by a real-time microscopy system to introduce reinforcement learning [11-13]. We demonstrate the solution of a standard problem of reinforcement learning - the navigation in a grid world. Due to the size of the microswimmer, noise introduced by Brownian motion if found to contribute considerably to both the learning process and the actions within a learned behavior. We extend the learning process to multiple swimmers and sharing of information. Our work represents a first step towards the integration of learning strategies into microsystems and provides a platform for the study of the emergence of adaptive and collective behavior.

References in corpus (15)

• Novel type of phase transition in a system of self-driven particles
• Active Particles in Complex and Crowded Environments
• Dynamical clustering and phase separation in suspensions of self-propelled colloidal particles
• Active Motion of Janus Particle by Self-thermophoresis in Defocused Laser Beam
• Active Brownian Motion Tunable by Light
• Gravitaxis of asymmetric self-propelled colloidal particles
• Targeted Assembly and Synchronization of Self-Spinning Microgears
• Flow Navigation by Smart Microswimmers via Reinforcement Learning
• Active Particles Bound by Information Flows
• Machine learning strategies for path-planning microswimmers in turbulent flows
• Optimal steering of a smart active particle
• Finding Efficient Swimming Strategies in a Three Dimensional Chaotic Flow by Reinforcement Learning
• Modelling collective motion based on the principle of agency
• Optimal decision making for sperm chemotaxis in the presence of noise
• Brownian Molecules Formed by Delayed Harmonic Interactions

Cited by in corpus (8)

• The 2019 Motile Active Matter Roadmap
• Zermelo's problem: Optimal point-to-point navigation in 2D turbulent flows using Reinforcement Learning
• Smart Inertial Particles
• Optimising low-Reynolds-number predation via optimal control and reinforcement learning
• Taming Polar Active Matter with Moving Substrates: Directed Transport and Counterpropagating Macrobands
• Reinforcement Learning reveals fundamental limits on the mixing of active particles
• Autotaxis of a Run-and-Tumble Agent via Reinforcement Learning
• Convolutional Neural Networks for Real-Time Localization and Classification in Feedback Digital Microscopy