Reinforcement Learning Enables Autonomous Microrobot Navigation and Intervention in Simulated Blood Capillaries

2026-06-23

AI Summary

Autonomous navigation and intervention of microrobots in complex capillary environments remain challenging due to the difficulties in accounting for realistic hemodynamics, vascular branching, and cellular obstacles. This work addresses these challenges by constructing a high-fidelity vascular simulation environment that integrates real blood flow fields, dynamic red blood cells, and anatomically accurate branching structures. By combining deep reinforcement learning with chemotactic control, the study achieves, for the first time, a generalizable navigation strategy that requires no retraining across diverse scenarios. The proposed approach successfully enables microrobots to perform targeted occlusion and recanalization tasks, restoring blood flow to healthy baseline levels. Furthermore, it reveals fundamental physical limits governing microrobot navigation dictated by size and speed, and uncovers universal behavioral patterns such as run-and-rotate and search-and-sit.

Abstract

Autonomous microrobots navigating biological vasculature could enable targeted drug delivery and thrombolysis, yet training control policies for realistic environments remains an open challenge. Prior reinforcement learning (RL) studies of microrobotic navigation have been limited to idealized geometries that omit complex hydrodynamic flow fields, confined branching structures, and dense cellular obstacles found in vivo. Here, we develop a physically grounded simulation of a blood capillary network, incorporating realistic hydrodynamic flow fields, explicit red blood cell dynamics, and anatomically derived branching geometry, and train deep RL agents to navigate it via chemotaxis. We systematically map the physical limits of navigation across robot size and swimming speed, revealing a forbidden regime where Brownian motion and flow overcome propulsion. Successful agents independently discover multiple universal strategy types, including run-and-rotate and energy-efficient search-and-sit policies, regardless of robot parameters. Without retraining, these agents perform targeted blocking and unblocking of capillary flow, restoring throughput to healthy baseline levels. These results establish RL as a viable framework for developing autonomous microrobotic intervention strategies in complex biological environments.

Problem

• microrobot navigation
• reinforcement learning
• blood capillaries
• autonomous intervention
• hydrodynamic flow

Innovation

• reinforcement learning
• microrobot navigation
• blood capillary simulation
• chemotaxis
• autonomous intervention