Cooperative Object Transportation using Gibbs Random Fields

2021 IEEE/RSJ International Conference on Intelligent Robots and Systems


Project maintained by rezeck Hosted on GitHub Pages — Theme by mattgraham

This paper presents a novel methodology that allows a swarm of robots to perform a cooperative transportation task. Our approach consists of modeling the swarm as a Gibbs Random Field (GRF), taking advantage of this framework's locality properties. By setting appropriate potential functions, robots can dynamically navigate, form groups, and perform cooperative transportation in a completely decentralized fashion. Moreover, these behaviors emerge from the local interactions without the need for explicit communication or coordination. To evaluate our methodology, we perform a series of simulations and proof-of-concept experiments in different scenarios. Our results show that the method is scalable, adaptable, and robust to failures and changes in the environment.

Citation

@inproceedings{rezeck2021collective,
  author={Rezeck, Paulo and Assunção, Renato M. and Chaimowicz, Luiz},
  booktitle={2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)}, 
  title={Cooperative Object Transportation using Gibbs Random Fields}, 
  year={2021},
  volume={},
  number={},
  pages={9131-9138},
  doi={10.1109/IROS51168.2021.9635928}
}

Usage

Prerequisites

Installation

Option 1: For Docker users

Option 2: System installation

Demo

Execution

Setup

roslaunch hero_gazebo gazebo_wizard.launch

Also set the number of robots in the grf controller param.yaml file:

# Number of robots in the environment
robots: 30
.
.
.

Edit the launch spawn_object.launch and select the object you want to spawn:

<!-- Select a object shape: "rectangular_prism, triangular_prism, polygonal_prism" -->
    <arg name="object_shape" default="rectangular_prism" />
    <param name="object_shape" value="$(arg object_shape)" />

Related Publications