Minute-Scale Training for Microrobot Navigation

2026-08-01

AI Summary

This work addresses the prohibitively long training times—ranging from hours to days—associated with conventional deep reinforcement learning approaches for microrobot navigation, which hinder rapid deployment and fine-tuning. To overcome this limitation, the authors propose an efficient learning framework that integrates fully vectorized, highly parallel simulation, a LiDAR-inspired perception model, dynamic feasibility checking, and a novel task-shaping regularization (TSR) reward mechanism. This approach reduces training time to under ten minutes, decreases action jitter by at least 33.7%, and improves obstacle avoidance success rates by at least 2.1%. Moreover, it achieves zero-shot generalization across unseen microrobot morphologies and environments, substantially accelerating policy convergence while enhancing robustness.

Abstract

Microrobots hold significant potential for various applications, where targeted navigation is a basic requirement. Deep reinforcement learning (DRL) has recently emerged as a powerful paradigm for fully autonomous microrobot navigation. Yet, current DRL-based approaches pay limited attention to learning efficiency and effectiveness, requiring hours to days for model training. Consequently, this impedes both rapid practical deployment and parameter optimization. To address these challenges, we present a learning framework that enables effective microrobot navigation policies to be trained within minutes. In the proposed framework, we develop a fully vectorized simulator with more than 10,000 artificial vascular environments, parallelizing dynamics, LiDAR-inspired perception, and feasibility checks across thousands of environments to achieve roughly 190,000 transitions per second. To achieve effectiveness in the fast training, we propose a task-shaping-regularization (TSR) reward framework. The TSR framework accelerates convergence, improves final performance, reduces action variation by at least 33.7%, and increases obstacle clearance by at least 2.1% across all evaluated scenarios. Results show that the proposed learning framework reduces training time to under 10 minutes, while supporting zero-shot deployment across distinct microrobot types and navigation scenarios. Collectively, this framework can substantially shorten the design loop and accelerate the deployment of autonomous microrobots.

Problem

• microrobot navigation
• deep reinforcement learning
• training efficiency
• learning effectiveness
• autonomous deployment

Innovation

• vectorized simulation
• task-shaping regularization
• minute-scale training
• microrobot navigation
• zero-shot deployment