End-to-End Low-Level Neural Control of an Industrial-Grade 6D Magnetic Levitation System

2025-09-01

AI Summary

To address the dynamic complexity of industrial-grade six-degree-of-freedom (6D) magnetic levitation systems and the limited generalizability and performance of conventional hand-crafted model-based controllers, this paper proposes the first end-to-end neural low-level controller. The controller directly maps raw sensor measurements and 6D reference pose commands to coil current references, entirely bypassing explicit physical modeling and hierarchical control architectures. Built upon an interactive, data-driven deep learning framework, it achieves high-precision positioning (sub-micrometer accuracy), strong robustness, and cross-operating-condition generalization on real hardware. Its core contribution lies in pioneering neural control at the current level—eliminating reliance on expert knowledge and precise system models—thereby significantly enhancing deployment flexibility and closed-loop performance. Code, trained models, and experimental videos are publicly released.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Intelligent Robots: Behavior Learning & Control
• Machine Learning: Deep Neural Architectures and Foundation Models

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Magnetic levitation is poised to revolutionize industrial automation by integrating flexible in-machine product transport and seamless manipulation. It is expected to become the standard drive for automated manufacturing. However, controlling such systems is inherently challenging due to their complex, unstable dynamics. Traditional control approaches, which rely on hand-crafted control engineering, typically yield robust but conservative solutions, with their performance closely tied to the expertise of the engineering team. In contrast, neural control learning presents a promising alternative. This paper presents the first neural controller for 6D magnetic levitation. Trained end-to-end on interaction data from a proprietary controller, it directly maps raw sensor data and 6D reference poses to coil current commands. The neural controller can effectively generalize to previously unseen situations while maintaining accurate and robust control. These results underscore the practical feasibility of learning-based neural control in complex physical systems and suggest a future where such a paradigm could enhance or even substitute traditional engineering approaches in demanding real-world applications. The trained neural controller, source code, and demonstration videos are publicly available at https://sites.google.com/view/neural-maglev.

Problem

• Neural controller for 6D magnetic levitation system control
• Overcoming complex unstable dynamics in industrial automation
• Replacing traditional hand-crafted control engineering approaches

Innovation

• End-to-end neural controller for 6D magnetic levitation
• Direct mapping from sensor data to coil commands
• Generalizes to unseen situations while maintaining robustness