Reinforcement learning of optimal active particle navigation

TL;DR

This paper introduces a deep reinforcement learning method to determine the asymptotically optimal navigation paths for self-propelled particles in complex environments, bypassing traditional analytical approaches.

Contribution

It presents the first reinforcement learning framework capable of finding optimal navigation paths for active particles without reward shaping or heuristics.

Findings

Successfully computes optimal paths in complex environments.

Demonstrates superiority over analytical methods.

Provides a universal path planning approach for active particles.

Abstract

The development of self-propelled particles at the micro- and the nanoscale has sparked a huge potential for future applications in active matter physics, microsurgery, and targeted drug delivery. However, while the latter applications provoke the quest on how to optimally navigate towards a target, such as e.g. a cancer cell, there is still no simple way known to determine the optimal route in sufficiently complex environments. Here we develop a machine learning-based approach that allows us, for the first time, to determine the asymptotically optimal path of a self-propelled agent which can freely steer in complex environments. Our method hinges on policy gradient-based deep reinforcement learning techniques and, crucially, does not require any reward shaping or heuristics. The presented method provides a powerful alternative to current analytical methods to calculate optimal trajectories and opens a route towards a universal path planner for future intelligent active particles.