Hybrid Image-Based Motion Tracking and Adaptive Polynomial Kalman Filter for a Bio-Inspired Dual-Sheath Needle System

Abstract

This preliminary study investigated sensor-fusion methods for accurate motion estimation in a bio-inspired dual-sheath needle system designed for biopsy applications. The needle system draws inspiration from the segmented dual-sheath mechanism of ovipositors found in Hymenoptera species, which allows controlled and precise movement. The aim is to develop a reliable tracking system for both the needle tip and base, which is essential for accurate needle placement during biopsy procedures. A hybrid tracking approach was used to achieve this, combining image-based tracking of the needle tip with an Inertial Measurement Unit (IMU) sensor for needle-base tracking. An Adaptive Polynomial Kalman Filter (APKF) was applied to improve the motion estimation accuracy. Performance tests on a linear-rail system demonstrated the ability of the developed system to provide accurate and consistent motion estimates. The results suggest that this bio-inspired needle system using sensor fusion and advanced filtering holds promise for use in medical robotics, potentially improving the precision of biopsy procedures.