Research Analyzer
← Back ICRA 2026

CareBot-H: Enhancing Patient Transfer with Biomimetic Design and Trajectory Deformation Algorithm

Deliang Zhu, Peizheng Li, Chunlei Meng, Jiexin Xie, Yang Li, Shijie Guo

PDF

AI summary

Key figure (auto-extracted from paper)
CareBot-H achieves safer, smoother, and more efficient patient transfers in confined spaces by combining biomimetic arms with a VAE-based trajectory prior and tactile-informed ZMP stability control.
Humanoid nursing robot Patient transfer Trajectory deformation Tactile sensing ZMP stability Biomimetic design

Problem

Existing nursing robots struggle with safe patient transfers in confined environments due to bulky designs, limited maneuverability, and insufficient tactile feedback for dynamic human-robot interaction.

Approach

The authors developed CareBot-H, a humanoid robot with biomimetic arms and distributed tactile sensors, controlled by a trajectory deformation algorithm that merges offline VAE-encoded expert motions with online tactile-informed ZMP adjustments for real-time stability.

Key results

  • Biomimetic dual-arm design with 850 mm reach and high torque-to-weight ratio
  • VAE-based offline learning captures anthropomorphic motion priors from teleoperation data
  • Online tactile-informed ZMP model enables real-time trajectory correction for dynamic stability
  • Experimental validation shows smoother, more efficient transfers with significantly reduced ZMP deviations versus manual teleoperation

Why it matters

Provides a scalable, safe automation solution for physically demanding patient transfers in compact nursing facilities and real-world elderly care settings.

Abstract

This paper introduces the CareBot-H Robot, a humanoid nursing robot designed to perform patient transfer tasks in confined environments. The robot is equipped with biomimetic arms that replicate human arm size and function, and distributed tactile sensors that enhance operational safety during physical contact. To achieve stable and humanoid motion, a trajectory deformation algorithm is proposed. The method comprises an offline phase, where expert demonstrations are encoded into prior trajectories using a Variational Autoencoder (VAE), and an online phase, where a tactile-informed Zero-Moment Point (ZMP) model enables real-time trajectory adjustment. Experimental validation with human participants demonstrates that the proposed approach outperforms manual teleoperation, producing smoother and more efficient transfer trajectories while significantly reducing deviations between actual and ideal ZMP. These results indicate that the CareBot-H achieves reliable and safe patient transfer performance, offering practical potential for deployment in real-world nursing care scenarios.

Index terms

Safety in HRI Touch in HRI Physical Human-Robot Interaction

Related papers