High Fidelity Capture, Reconstruction, and Transfer of Human Demonstrations for Robot-Assisted Bathing

2026-08-10

AI Summary

This work addresses the challenge of achieving safe and reliable continuous physical human–robot interaction in high-value clinical tasks such as assisted bathing. The authors propose a high-fidelity framework centered on contact regions, through which they construct the first high-quality human–human bathing interaction dataset encompassing synchronized motion, deformation, contact, and force information. Leveraging this dataset, they demonstrate multi-level transfer control from human demonstrations to a dexterous soft robotic hand. By integrating high-precision motion and tactile sensing, both open-loop and closed-loop control strategies are developed to drive an arm-mounted soft hand, successfully reproducing bathing motions on a human mannequin and validating the approach’s efficacy. All data and code will be publicly released to advance research in physical human–robot interaction.

Abstract

Despite the demand for robots in high-value clinical tasks like bathing, contemporary systems still lack the safety and reliability required for complex, sustained physical interaction with humans. A key challenge hindering the development of such systems is that collecting, understanding, and effectively transferring highly dynamic, contact-rich human bathing demonstrations is difficult, even with modern motion and tactile sensing equipment. We present a straightforward, but effective framework for doing so with high fidelity by utilizing contact regions as a key processing primitive. We use our framework to build a dataset of bathing demonstrations performed by trained clinicians on human subjects. We then use this dataset to design and control an arm-mounted dexterous soft hand to perform bathing tasks on a mannequin using open- and closed-loop strategies. Our dataset is the first to provide high quality synchronized motion, shape, contact, and force during sustained, contact-rich human-human interaction, and our transfer strategies demonstrate effective use of these data across multiple levels of the robotics stack. All relevant materials will be publicly released to enable further advancements in physical human-robot interaction (pHRI) research.

Problem

• robot-assisted bathing
• physical human-robot interaction
• contact-rich demonstrations
• human demonstration transfer
• high-fidelity capture

Innovation

• contact regions
• high-fidelity demonstration transfer
• soft robotic hand
• synchronized multimodal dataset
• physical human-robot interaction