GEBREZGIHER, MICHEALE HAILESLASSIE (Chonnam National University), Choi, Jiyoung (Chonnam National University), Hong, Ayoung (Chonnam National University)

Tracking Deformable Linear Objects under Occlusion Using Cubic Spline Interpolation and Unscented Kalman Filter

Scheduled for presentation during the Regular Session "Special Topics in Robotic Systems" (WeCT5), Wednesday, July 16, 2025, 17:50−18:10, Room 109

Joint 10th IFAC Symposium on Mechatronic Systems and 14th Symposium on Robotics, July 15-18, 2025, Paris, France

This information is tentative and subject to change. Compiled on July 16, 2025

Abstract

This paper presents an online state estimation method for Deformable Linear Objects (DLOs) under occluded views using the Unscented Kalman Filter (UKF). We integrate a stereo vision-based tracking algorithm for data acquisition with a data-driven dynamic model to predict the 3D shape of the DLO online. When the DLOs are partially occluded or out of view, cubic spline interpolation is used to estimate the missing nodes. To validate the proposed method, we performed experiments with two robotic manipulators holding both ends of the cable and achieved an average tracking error of under 4 mm.