A Learning-Based Approach for Contact Detection, Localization, and Force Estimation of Continuum Manipulators With Integrated OFDR Optical Fiber

Mobina Tavangarifard; Jonathan S. Kacines; Qiyu Li; Farshid Alambeigi

cs.ROMar 12, 2026

A Learning-Based Approach for Contact Detection, Localization, and Force Estimation of Continuum Manipulators With Integrated OFDR Optical Fiber

Authors: Mobina Tavangarifard, Jonathan S. Kacines, Qiyu Li, Farshid Alambeigi

Organizations: Walker Department of Mechanical Engineering and Texas Robotics, The University of Texas at Austin, Austin, TX 78712, USA. · Department of Computer Science, The University of Texas at Austin, Austin, TX 78712, USA.

Abstract

Continuum manipulators (CMs) are widely used in minimally invasive procedures due to their compliant structure and ability to navigate deep and confined anatomical environments. However, their distributed deformation makes force sensing, contact detection, localization, and force estimation challenging, particularly when interactions occur at unknown arc-length locations along the robot. To address this problem, we propose a cascade learning-based framework (CLF) for CMs instrumented with a single distributed Optical Frequency Domain Reflectometry (OFDR) fiber embedded along one side of the robot. The OFDR sensor provides dense strain measurements along the manipulator backbone, capturing strain perturbations caused by external interactions. The proposed CLF first detects contact using a Gradient Boosting classifier and then estimates contact location and interaction force magnitude using a CNN--FiLM model that predicts a spatial force distribution along the manipulator. Experimental validation on a sensorized tendon-driven CM in an obstructed environment demonstrates that a single distributed OFDR fiber provides sufficient information to jointly infer contact occurrence, location, and force in continuum manipulators.

Figures & tables

Fig. 1: A) Conceptual illustration of a continuum manipulator interacting with an obstacle at an unknown location along its geometry. The interaction force and contact location are measured using a force gauge and a motion capture system. An embedded OFDR optical fiber measures the distributed strain along the manipulator. B) Physical experimental setup used to evaluate the proposed CLF. The figure shows the shape-sensing assembly and its integration inside the continuum manipulator (CM).

Fig. 2: The proposed CLF architecture for contact detection, localization, and magnitude of force estimation.

Fig. 3: Gradient Boosting Classifier for Contact Detection. Figure illustrates the cascade process using weighted strain data for contact classification.

Fig. 4: Illustration of the CNN architecture with FiLM generator, processing strain signals and motor position for force localization.

Block	In Ch.	Out Ch.	Kernel	Padding	Dilation
Block 1	1	16	7	3	2
Block 2	16	32	7	3	2
Block 3	32	64	3	1	1

TABLE I: Convolutional Block Architecture

Layer	In Features	Out Features	Purpose
Linear 1	$N_{\text{scalar}}$	128	Feature projection
Linear 2	128	128	Output $\gamma$ and $\beta$

TABLE II: FiLM Generator Architecture

Layer	In Features	Out Features	Activation
Linear 1	$Dim_{\text{FiLM}}$	128	ReLU
Linear 2	128	64	ReLU
Linear 3	64	$N_{\text{out}}$	–

TABLE III: Fully Connected Layer Architecture

Fig. 5: Bar plots showing per-test performance for contact classification, including ROC-AUC, Precision, and Recall scores across different test IDs.

Fig. 6: Force and localization absolute error distributions. Mean absolute error (MAE) indicated by horizontal line.

Label	ROC -AUC	Recall	Precision	Force MAE (N)	Localization MAE (mm)
$\mathrm{cv}_{1}$	0.995	0.943	0.957	0.013	2.936
$\mathrm{cv}_{2}$	0.996	0.989	0.913	0.008	0.494
$\mathrm{cv}_{3}$	0.963	0.692	1.000	0.017	0.646
$\mathrm{cv}_{4}$	0.983	0.824	0.913	0.008	6.477
$\mathrm{cv}_{5}$	0.990	0.745	1.000	0.009	0.497
$\mathrm{cv}_{6}$	0.927	0.570	0.880	0.014	0.728

TABLE IV: Contact Classification Performance Metrics, Force and Localization Errors

Explore similar work

Cable-driven Continuum Robotics: Proprioception via Proximal-integrated Force Sensing
Mar 8, 2026Gang Zhang, Junyan Yan, Jibiao Chen +1Robotic Perception Tactile Sensing
Department of Mechanical and Automation Engineering and T Stone Robotics Institute, The Chinese University of Hong Kong, Hong Kong · Curve Robotics Limited, Hong Kong · Shun Hing Institute of Advanced Engineering and Multi-Scale Medical Robotics Center, Hong Kong
Multi-Contact Force Estimation for Continuum Robots via Gaussian-Parameterized Factor Graphs
Jun 28, 2026Aditya Prakash, Panagiotis TsiotrasRobot State Estimation
School of Aerospace Engineering Georgia Institute of Technology Atlanta, GA, USA
Tactile Reconstruction of Contact Task Frames and Forces for Hybrid Force/Motion Control
Oct 7, 2026Antonio Rapuano, Simone Orelli, Barbara Bazzana +2Force-Controlled Robotic Manipulation Tactile Sensing
Department of Computer, Control and Management Engineering, Sapienza University of Rome, 00185 Rome, Italy. · Robotics and Mechatronics Department, Faculty of Electrical Engineering, Mathematics and Computer Science, University of Twente, 7500 AE Enschede, The Netherlands.

Mar 8, 2026J/K move · Enter open · S save

Cable-driven Continuum Robotics: Proprioception via Proximal-integrated Force Sensing

Gang Zhang, Junyan Yan, Jibiao Chen, Shing Shin Cheng

Department of Mechanical and Automation Engineering and T Stone Robotics Institute, The Chinese University of Hong Kong, Hong Kong · Curve Robotics Limited, Hong Kong · Shun Hing Institute of Advanced Engineering and Multi-Scale Medical Robotics Center, Hong Kong

Robotic Perception Tactile Sensing Soft Robotics

Open paper AskPDF arXiv

Micro-scale continuum robots face significant limitations in achieving three-dimensional contact force perception, primarily due to structural miniaturization, nonlinear mechanical, and sensor integration. To overcome these limitations, this paper introduces a novel proprioception method for cable-driven continuum robots based on proximal-integrated force sensing (i.e., cable tension and six-axis force/torque (F/T) sensor), inspired by the tendon-joint collaborative sensing mechanism of the finger. By integrating biomechanically inspired design principles with nonlinear modeling, the proposed method addresses the challenge of force perception (including the three-dimensional contact force and the location of the contact point) and shape estimation in micro-scale continuum robots. First, a quasi-bionic mapping between human tissues/organs and robot components is established, enabling the transfer of the integrated sensing strategy of tendons, joints, and neural feedback to the robotic system. Second, a multimodal perception strategy is developed based on the structural constraints inherent to continuum robots. The complex relationships among mechanical and material nonlinearities, robot motion states, and contact forces are formulated as an optimization problem to reduce the perception complexity. Finally, experimental validation demonstrates the effectiveness of the proposed method. This work lays the foundation for developing safer and smarter continuum robots, enabling broader clinical adoption in complex environments.

Abstract

Explore similar work

Cable-driven Continuum Robotics: Proprioception via Proximal-integrated Force Sensing

Multi-Contact Force Estimation for Continuum Robots via Gaussian-Parameterized Factor Graphs

Tactile Reconstruction of Contact Task Frames and Forces for Hybrid Force/Motion Control