Enhanced localization strategy for magnetic capsule robot using on-board nine-axis IMU through incorporation of alternating magnetic field
Capsule robots have gained increasing prominence in gastroscopy for their diagnostic capabilities and non-invasive accessibility to narrow passages with reduced risks. Magnetic positioning plays a crucial role in mapping the trajectory of the capsule robot and targeting specific areas of interest. H...
Gespeichert in:
Veröffentlicht in: | IEEE transactions on instrumentation and measurement 2024-01, Vol.73, p.1-1 |
---|---|
Hauptverfasser: | , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Capsule robots have gained increasing prominence in gastroscopy for their diagnostic capabilities and non-invasive accessibility to narrow passages with reduced risks. Magnetic positioning plays a crucial role in mapping the trajectory of the capsule robot and targeting specific areas of interest. However, the current magnetic positioning technology is limited by challenges of incompatibility with external magnetic manipulation systems, power and size constraints, and reduced resolution caused by accumulated errors during attitude estimation. Here, we propose an enhanced localization strategy for magnetic capsule robots with an on-board nine-axis inertial measurement unit (IMU) through the incorporation of an alternating magnetic field. Motion control of the capsule robot is achieved by an extracorporeal robotically manipulated permanent magnet, which also serves as a magnetic reference for determining the capsule robot's relative position. An alternating magnetic field is applied using a single-axis Helmholtz coil to compensate for the accumulated error in the yaw angle of the IMU. A digital lock-in amplifier algorithm is employed to decouple the static magnetic field generated by the permanent magnet from the alternating magnetic fields. To improve the accuracy of attitude and position estimation, we introduce an improved magnetic dipole model, which is optimized through high-resolution spatial mapping of the magnetic field of the permanent magnet. Experimental results demonstrate that within a rectangular space of 100 mm × 100 mm, our proposed strategy achieves an average position error of 2.52 mm and an average angle error of 1.53°, meeting the accuracy requirements for clinical examinations. |
---|---|
ISSN: | 0018-9456 1557-9662 |
DOI: | 10.1109/TIM.2024.3350144 |