ROS-LLM:

A ROS framework for embodied AI with task feedback and structured reasoning

Christopher E. Mower1    Yuhui Wan1,2, Hongzhan Yu1, Antoine Grosnit1,3   
Jonas Gonzalez-Billandon1, Matthieu Zimmer1, Jinlong Wang4, Xinyu Zhang4,
Yao Zhao5, Anbang Zhai5, Puze Liu3, Daniel Palenicek3, Davide Tateo3, Cesar Cadena6,
Marco Hutter6, Jan Peters3    Guangjian Tian7, Yuzheng Zhuang8, Kun Shao1,
Xingyue Quan8, Jianye Hao8, Jun Wang2    Haitham Bou-Ammar1

1Huawei Noah's Ark Lab, London, UK.

2University of Leeds, Leeds, UK.

3Technical University of Darmstadt, Darmstadt, Germany.

4East China Normal University, Shanghai, China.

5Huawei Technologies, Hangzhou, China.

6ETH Zurich, Zurich, Switzerland.

7Huawei Noah's Ark Lab, Hong Kong, China.

8Huawei Noah's Ark Lab, Shenzhen, China.

9University College London, London, UK.

  
Paper arXiv Code
We present a framework for intuitive robot programming by non-experts, leveraging natural language prompts and contextual information from the Robot Operating System (ROS).

Framework

frameworkImage

BibTeX

@misc{mower2024rosllmrosframeworkembodied,
      title={ROS-LLM: A ROS framework for embodied AI with task feedback and structured reasoning}, 
      author={Christopher E. Mower and Yuhui Wan and Hongzhan Yu and Antoine Grosnit and Jonas Gonzalez-Billandon and Matthieu Zimmer and Jinlong Wang and Xinyu Zhang and Yao Zhao and Anbang Zhai and Puze Liu and Davide Tateo and Cesar Cadena and Marco Hutter and Jan Peters and Guangjian Tian and Yuzheng Zhuang and Kun Shao and Xingyue Quan and Jianye Hao and Jun Wang and Haitham Bou-Ammar},
      year={2024},
      eprint={2406.19741},
      archivePrefix={arXiv},
      primaryClass={cs.RO},
      url={https://arxiv.org/abs/2406.19741}, 
}