ICRA 2026: Collaborative Human-Robot Object Transportation Using a Deformable Sheet

Weijian Zhang, Charlie Street, Masoumeh Mansouri

University of Birmingham

Overview

We tackle real-time formation trajectory planning for collaborative object transportation in complex environments using a team of nonholonomic robots and a human.

Features

• A multi-modal real-time formation trajectory planning framework where leadership switches between the human and robots to maximize safety and transportation efficiency.

• A method for computing feasible regions that can be visualized to the human to guide them during obstacle crossing.

Requirements

• ROS Noetic or later
• Ubuntu 20.04 or later
• yaml-cpp 0.8.0 or later

Installation

# Create a new workspace:
mkdir -p ~/HRCTF/src
cd ~/HRCTF/src
catkin_init_workspace

# Clone the package into the workspace
git clone git@github.com:HyPAIR/HRCTF.git

# Install dependencies
cd ~/HRCTF
rosdep install --from-paths src --ignore-src --rosdistro noetic -y \
  --skip-keys="lightsfm ecbs"

Quick Start

Before running the code, make sure that the joystick is properly connected to your PC!

# Build
cd ~/HRCTF
catkin_make

# Terminal 1 - Launch simulation
source devel/setup.bash
roslaunch liom_local_planner heterogeneous_triangle_1h2r.launch 

# Terminal 2 - Launch formation planner
source devel/setup.bash
roslaunch liom_local_planner intention_driven_test.launch 

Configuration

Edit parameters in:

• src/liom_local_planner/config/planner/params.yaml

Video

A simulation and real-world experiments video demonstrating our proposed framework can be found at bilibili/youtube.

Citation

If you find this work useful, please cite (paper):

@inproceedings{zhang2026hrctf,
  title={Collaborative Human-Robot Object Transportation Using a Deformable Sheet},
  author={Zhang, Weijian and Street, Charlie and Mansouri, Masoumeh},
  booktitle={2026 IEEE International Conference on Robotics and Automation},
  year={2026},
  organization={IEEE}
}

Acknowledgement

We use datmo to track the human and estimate its position, orientation, and velocity. We use DecompROS for safe corridor generation and visualization.