ROSGPT is a pioneering approach that combines the power of ChatGPT and ROS (Robot Operating System) to redefine human-robot interaction. By leveraging large language models like ChatGPT, ROSGPT enables the conversion of unstructured human language into actionable robotic commands. This repository contains the implementation of ROSGPT, allowing developers to explore and contribute to the project.
Author: Anis Koubaa
Citation: Koubaa, A. (2023). ROSGPT: Next-Generation Human-Robot Interaction with ChatGPT and ROS. Preprints.org, 2023, 2023040827. https://www.preprints.org/manuscript/202304.0827/v2
BibTeX Citation:
@article{koubaa2023rosgpt,
title={ROSGPT: Next-Generation Human-Robot Interaction with ChatGPT and ROS},
author={Koubaa, Anis},
journal={Preprints.org},
year={2023},
volume={2023},
pages={2023040827},
doi={10.20944/preprints202304.0827.v2}
}
Explore ROSGPT in action with this video demonstration, showcasing the process of getting started and the capabilities of the system.
The ROSGPT ROS2 package includes a collection of scripts that work together to provide a convenient way of translating natural human language text into structured JSON commands, which can be utilized by robots like Turtlesim and Turtlebot3. Below is a brief overview of each script:
/voice_cmd
topic.ros2 run
command to execute this node.rosgpt_client_node.py
, this script sends POST requests with natural human language text to the ROSGPT REST server, but without implementing a ROS2 node. It solely functions as a REST client for ROSGPT. Use the python
command, not ros2 run
, to execute this script./voice_cmd
topic and receives JSON commands. The node parses the JSON command and determines the ROS2 primitives required to execute the specified tasks. In this script, a simple navigation task for the Turtlesim robot is considered, including move and rotate functions./voice_cmd
topic and receiving JSON commands. The JSON commands are parsed and transformed into navigation goal tasks for the Turtlebot3 robot.To get started with ROSGPT, follow these steps:
ros2 run rosgpt rosgpt
ros2 run turtlesim turtlesim_node
ros2 run rosgpt rosgptparser_turtlesim
ros2 run rosgpt rosgpt_client_node
This ROS 2 package was tested using ROS 2 Humble with Ubuntu 22.04. It should also work with ROS 2 Foxy and other ROS 2 versions. You need to install the following dependencies:
Add your OpenAI API Key in your .bashrc
as an environment variable
echo 'export OPENAI_API_KEY=your_api_key' >> ~/.bashrc
Install the dependencies required for the text-to-speech functionality
sudo apt-get install libespeak1
sudo apt install ros-humble-turtlesim*
For ros2 Humble verison (Ubuntu 22.04) downgrading the setuptools is required
pip3 install --upgrade setuptools==58.0.2
Install python dependencies
cd rosgpt
pip3 install -r requirements.txt
then build the package
colcon build --packages-select rosgpt
source the workspace
source install/setup.bash
To get started with ROSGPT, got this section Getting Started
The ROSGPT REST API is a convenient way of interacting with ROSGPT. It allows you to send POST requests with natural human language text to the ROSGPT server, which will then translate the text into structured JSON commands. The JSON commands can be used to control robots like Turtlesim and Turtlebot3.
To use the ROSGPT REST API, follow these steps:
ros2 run rosgpt rosgpt
.ros2 run turtlesim turtlesim_node
.ros2 run rosgpt rosgptparser_turtlesim
.python rosgpt_client.py
.curl -X POST -H "Content-Type: application/json" -d '{"text":"move forward"}' http://localhost:5000/rosgpt
. You can replace "move forward" with any natural human language text you want. The ROSGPT server will translate the text into structured JSON commands and send them back to the client.ROS1 is an earlier version of the Robot Operating System (ROS), which is still widely used in many robotics applications. While ROSGPT was originally developed for ROS 2 Humble on Ubuntu 22.04, we recognize the importance of supporting ROS1 as well.
To use ROSGPT with ROS1, you will need to modify the ROS 2 code in the scripts to the corresponding ROS 1 code. We are actively working on developing this functionality, but it is still a work-in-progress.
If you have already developed an extension to enable ROSGPT to work with ROS1, we would love to hear from you! Please create a pull request in a new branch and we will review it for inclusion in the ROSGPT repository.
This project is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License. You are free to use, share, and adapt this material for non-commercial purposes, as long as you provide attribution to the original author(s) and the source.
As this project is still under progress, contributions are welcome! To contribute, please follow these steps:
Before submitting your pull request, please ensure that your changes do not break the build and adhere to the project's coding style.
For any questions or suggestions, please open an issue on the GitHub issue tracker.