A Mobile Magnetic Manipulation Platform for Gastrointestinal Navigation with Deep Reinforcement Learning Control

2026-01-22

AI Summary

This work addresses the limitations of conventional magnetic actuation systems for gastrointestinal targeted drug delivery, which rely on complex physical models for calibration, resulting in low deployment efficiency and restricted workspace. The authors propose the first model-free magnetic control framework based on deep reinforcement learning, integrating a UR5 collaborative robot with a four-electromagnet array. By employing the Soft Actor-Critic (SAC) algorithm and a simulation-to-reality transfer strategy, the system achieves rapid deployment without pre-calibration and enables precise capsule manipulation. The platform can be set up within 15 minutes and accurately controls a 7 mm magnetic capsule within a clinically relevant 30 cm × 20 cm workspace, achieving root-mean-square tracking errors of 1.18 mm and 1.50 mm for square and circular trajectories, respectively.

Technology Category

• Intelligent Robots: Manipulation
• Multiagent Systems: Adversarial Agents
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Targeted drug delivery in the gastrointestinal (GI) tract using magnetic robots offers a promising alternative to systemic treatments. However, controlling these robots is a major challenge. Stationary magnetic systems have a limited workspace, while mobile systems (e.g., coils on a robotic arm) suffer from a"model-calibration bottleneck", requiring complex, pre-calibrated physical models that are time-consuming to create and computationally expensive. This paper presents a compact, low-cost mobile magnetic manipulation platform that overcomes this limitation using Deep Reinforcement Learning (DRL). Our system features a compact four-electromagnet array mounted on a UR5 collaborative robot. A Soft Actor-Critic (SAC)-based control strategy is trained through a sim-to-real pipeline, enabling effective policy deployment within 15 minutes and significantly reducing setup time. We validated the platform by controlling a 7-mm magnetic capsule along 2D trajectories. Our DRL-based controller achieved a root-mean-square error (RMSE) of 1.18~mm for a square path and 1.50~mm for a circular path. We also demonstrated successful tracking over a clinically relevant, 30 cm * 20 cm workspace. This work demonstrates a rapidly deployable, model-free control framework capable of precise magnetic manipulation in a large workspace,validated using a 2D GI phantom.

Problem

• magnetic manipulation
• gastrointestinal navigation
• model-calibration bottleneck
• targeted drug delivery
• mobile magnetic control

Innovation

• Deep Reinforcement Learning
• Magnetic Manipulation
• Model-Free Control
• Sim-to-Real Transfer
• Soft Actor-Critic