Unified Magnetic 5-DoF Localization Framework for Capsule Robots Via PMMN-DBO: From Single to Multi-Robot Scenarios with Real-Time Control�Localization Co-Design
Zijin Zeng, Chan Li, Zaiyang Chen, Shunxiao Huang, Wenyan Niu, Hongyan Sun, Menglu Tan, Yingjian Guo, Lin Feng
AI summary
Problem
Clinical magnetic localization for capsule robots struggles with coupling between control and localization fields, complex near-field modeling, and real-time optimization in multi-capsule scenarios, leading to accuracy loss and instability.
Approach
A unified framework combining layered multi-source magnetic field modeling, online external-field compensation, and a novel parallel multi-task dung beetle optimizer with mirrored boundary reflection and elite mutation for global pose inversion.
Key results
- 0.59 mm/0.69° mean error with 20.2 ms computation for single-capsule
- Stable multi-capsule localization with 1.28 mm/1.13° (two) and 2.56 mm/2.83° (three) errors
- 1.33 mm/1.85° trajectory-tracking error under synchronized control-localization
- Superior accuracy and stability over conventional optimizers while maintaining real-time performance
Why it matters
Provides a robust, hardware-agnostic foundation for clinical-grade, closed-loop magnetic navigation and therapy in gastrointestinal capsule robotics.
Abstract
Motivated by clinical needs for precise navigation and safety, low-latency and high-precision localization has become a key enabler for capsule robots. A unified magnetic 5-DoF high-precision localization framework for capsule robots is presented. Building on layered multi-source magnetic field modeling, online external-field compensation, and global optimization-based inversion, the framework achieves real-time decoupling between control and localization fields, while providing a unified interface compatible with diverse hardware configurations and operation modes. On this basis, the PMMN-DBO algorithm is proposed, delivering high-accuracy and efficient localization in single- and multi-capsule scenarios, and supports synchronized control–localization. Experimentally, for single-capsule localization, mean errors are 0.59 mm/0.69° with a 20.2 ms computation time, surpassing conventional methods. In multi-capsule settings, localization errors remain low with stable convergence: mean errors are 1.28 mm/1.13° for two capsules and 2.56 mm/2.83° for three capsules. Under synchronized control–localization, trajectory-tracking errors reach 1.33 mm/1.85°. Overall, the proposed framework is unified, high-precision, efficient, and flexible, laying a general and reusable foundation for clinical-grade precise navigation and closed-loop magnetic control.