Towards Autonomous Tape Handling for Robotic Wound Redressing

TL;DR

This work tackles autonomous wound care by focusing on adhesive tape manipulation, a foundational subtask in wound redressing. It combines force-feedback imitation learning to master tape initial detachment with a differentiable-simulation–driven trajectory optimization for wrinkle-free tape placement, validated across phantom, arm, and anatomical skin models. The approach yields safer detachment with higher effectiveness and robust, wrinkle-free adhesion on diverse surfaces, demonstrated both in isolated subtasks and a full end-to-end pipeline. The findings suggest a viable path toward practical home wound care automation, with future directions in robust perception, grasping, and higher-level planning for broader wound locations.

Abstract

Chronic wounds, such as diabetic, pressure, and venous ulcers, affect over 6.5 million patients in the United States alone and generate an annual cost exceeding \$25 billion. Despite this burden, chronic wound care remains a routine yet manual process performed exclusively by trained clinicians due to its critical safety demands. We envision a future in which robotics and automation support wound care to lower costs and enhance patient outcomes. This paper introduces an autonomous framework for one of the most fundamental yet challenging subtasks in wound redressing: adhesive tape manipulation. Specifically, we address two critical capabilities: tape initial detachment (TID) and secure tape placement. To handle the complex adhesive dynamics of detachment, we propose a force-feedback imitation learning approach trained from human teleoperation demonstrations. For tape placement, we develop a numerical trajectory optimization method based to ensure smooth adhesion and wrinkle-free application across diverse anatomical surfaces. We validate these methods through extensive experiments, demonstrating reliable performance in both quantitative evaluations and integrated wound redressing pipelines. Our results establish tape manipulation as an essential step toward practical robotic wound care automation.

Figures (6)

• Figure 1: This work enables advanced tape manipulation skills such as tape initial detachment (TID), and tape placement. They help automating three critical steps for wound redressing, including dressing removal, tape preparation, and tape placement to secure secondary dressings.
• Figure 2: The tape initial detachment framework includes (a) demonstration data collection via human teleoperation and (b) imitation learning for training and inference. An operator uses a haptic device to control the robot tip, receiving force feedback from a sensor between the hand and arm, while in-hand cameras capture images. The network is trained on segmented image and force data to predict future tip movements or forces, with semantic segmentation initialized from a small set of labeled samples.
• Figure 3: Key geometric concepts that the proposed tape placing method considers. (a) illustrates a tape element's position and orientation. In each numerical step, (b) the algorithm finds the next free tape element's closest point $\hat{p}_i$ and its surface normal $n^S_{\hat{p}_i}$. (c) Then it computes an rotation axis $\mathbf{v}_i$ with cross product between the normal vector at the latest attachment element $n^S_{p_{i-1}}$ and $n^S_{\hat{p}_i}$. (d) In next step, the tape region is rotated around $\mathbf{v}_i$ to approach the surface.
• Figure 4: Visualization of the quantitative evaluation metrics used in this work. Left: effective removal length (ERL) is measured by marking the furthest detached tape position from the tape end. Right: tape placement security is quantified by recording the maximum reading that a force gauge (N) when forcefully removing the tape and gauze.
• Figure 5: Relationship between the resulting ERL and number of attempts. Within 3 attempts, ERL of the proposed TID method linearly increases as the number of attempts increases in multiple scenarios, showing that its capability of improving outcome with repetition