Session: 07-10-02: Medical Robotics, Rehabilitation, and Surgery

Paper Number: 193277

Kinematic Control for Robot-Assisted Epidural Needle Insertion Using Estimated Axial Force Feedback 

Epidural needle insertion is a medical procedure that requires precise control of the needle trajectory and careful interpretation of the interaction forces between the instrument and biological tissues. During manual execution, clinicians rely on changes in mechanical resistance to identify transitions between different anatomical layers. This perception of force is a critical component of the procedure and is typically acquired through clinical experience. As a result, studying and reproducing these interaction dynamics in a controlled environment represents an important challenge for both research and medical training. In this context, robotic systems provide a suitable experimental platform for investigating the mechanical interaction between surgical tools and tissue-like materials under repeatable and measurable conditions.

This research proposes a robot-assisted insertion approach based on kinematic control with axial force feedback estimated from the internal variables of the robotic system. Unlike methods that rely on external force sensors or complex multimodal perception systems, the proposed approach estimates the interaction force at the end effector using the torques measured at the robot joints and their relationship with the Cartesian space through the manipulator Jacobian. This strategy enables the estimation of the axial force exerted during insertion using only information available from the robot actuators, making it particularly suitable for robotic platforms that do not incorporate dedicated force sensors.

The proposed system operates under a kinematic control scheme in which the insertion trajectory is defined as a rectilinear motion along the intended insertion axis. During the execution of this motion, the system continuously monitors the joint torques and computes an estimate of the axial force applied by the end effector. This estimated force is then used as a feedback signal to regulate the insertion process. Specifically, the system can adjust the insertion velocity or stop the motion when the estimated axial force exceeds a predefined threshold, thereby limiting excessive interaction forces during contact with the medium.

Experimental validation is performed by executing controlled insertion trajectories into materials designed to emulate the mechanical behavior of tissues encountered during epidural procedures. These materials produce variations in mechanical resistance as the needle advances, generating characteristic force profiles that can be captured through the estimated interaction force derived from the robot’s internal measurements.

The main objective of this study is to evaluate the feasibility of using estimated axial force as a feedback variable for regulating the insertion process through kinematic control. By relying solely on actuator measurements, the proposed approach seeks to demonstrate that meaningful interaction information can be extracted without the need for dedicated force sensing hardware. It is expected that this method will allow the reproduction of insertion profiles consistent with the mechanical properties of the simulated medium, contributing to the development of accessible robotic platforms for the study of insertion-based medical procedures and for the investigation of interaction control strategies based on force estimation.

Presenting Author: Eduardo Flores-Rivas Universidad Nacional Autónoma de México, Facultad de Ingeniería

Presenting Author Biography: Eduardo Flores Rivas is a Master’s student in Mechanical Engineering in the Graduate Program in Engineering at the National Autonomous University of Mexico. His research interests include medical robotics, robot–tissue interaction, and force estimation methods for insertion-based procedures. His current work focuses on the development of control strategies for robot-assisted epidural needle insertion using axial force estimation derived from joint torque measurements. He also serves as a lecturer in basic engineering sciences at the Faculty of Engineering.

Authors:

Daniel Haro-Mendoza Universidad Nacional Autónoma de México, Facultad de Ingeniería
Eduardo Flores-Rivas Universidad Nacional Autónoma de México, Facultad de Ingeniería
Victor J. Gonzalez-Villela Universidad Nacional Autónoma de México, Facultad de Ingeniería
Daniel Padilla-Ramos Universidad Nacional Autónoma de México