Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

How can LLMs transform the robotic design process?

Nature Machine Intelligence volume 5, pages 561–564 (2023)Cite this article

Subjects

We show that large language models (LLMs), such as ChatGPT, can guide the robot design process, on both the conceptual and technical level, and we propose new human–AI co-design strategies and their societal implications.

Large language models (LLMs)1 will fundamentally change the robotics landscape by providing robots with the unprecedented capability to understand and analyse natural language. The key advantage of LLMs is their ability to process and internalize large amounts of text data such as instructions, technical manuals and academic articles, and to factually and coherently respond to questions using this implicit knowledge. The potential of utilizing these powerful AI tools within robotics has already been shown through their ability to synthesize code from text prompts2 and to translate natural-language instructions into actions executable by robots3,4. However, recent improvements in the availability and capabilities of LLMs opens new opportunities, and they may now contribute to another bottleneck of robotics, design. Leveraging their emerging capabilities5,6, LLMs can deliver a dialogue that educates, stimulates and guides humans in building robots. These capabilities could fundamentally change the methodology by which we design robots, changing the role of humans and enriching and simplifying the design process. So, how can LLMs transform the robotic design process, and what are the associated opportunities and risks?

This is a preview of subscription content, access via your institution

Access options

Rent or buy this article

Prices vary by article type

from$1.95

to$39.95

Learn more

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Design pipeline.
Fig. 2: An AI model designed this robotic gripper.
Fig. 3: Opportunities and risks of human–LLM design collaboration.

References

  1. Brants, T., Popat, A. C., Xu, P., Och, F. J. & Dean, J. In Proc. 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning (EMNLP-CoNLL) 858–867 (ACM, 2007).

  2. Budzianowski, P. & Vulić, I. Preprint at https://arxiv.org/1907.05774 (2019).

  3. Brohan, A. et al. In Conference on Robot Learning 287–318 (PMLR, 2023).

  4. Huang, W., Abbeel, P., Pathak, D. & Mordatch, I. Preprint at https://arxiv.org/abs/2201.07207 (2022).

  5. Wei, J. et al. Preprint at https://doi.org/10.48550/arXiv.2206.07682 (2022).

  6. Floridi, L. & Chiriatti, M. Minds Mach. 30(4), 681–694 (2020).

    Article  Google Scholar 

  7. Chen, M. et al. Preprint at https://doi.org/10.48550/arXiv.2107.03374 (2021).

  8. Wolfram, S. Stephen Wolfram Writings https://go.nature.com/3ozHgNf (2023).

  9. Ramesh, A. et al. Preprint at https://doi.org/10.48550/arXiv.2204.06125 (2022).

  10. Shanahan, M. Preprint at https://doi.org/10.48550/arXiv.2212.03551 (2022).

  11. Stokel-Walker, C. Nature 613, 620–621 (2023).

    Article  Google Scholar 

  12. Vincent, J. The Verge https://go.nature.com/3N2HTIB (2023).

  13. George, A. & Walsh, T. Nat. Mach. Intell. 4, 1057–1060 (2022).

    Article  Google Scholar 

  14. Anonymous. Nat. Mach. Intell. 5, 1 (2023).

    Article  Google Scholar 

  15. Stella, F. & Hughes, J. Front. Robot. AI 9, 1059026 (2023).

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. CREATE Lab, EPFL, Lausanne, Switzerland

    Francesco Stella & Josie Hughes

  2. Department of Cognitive Robotics, Delft University of Technology, Delft, the Netherlands

    Francesco Stella & Cosimo Della Santina

  3. Institute of Robotics and Mechatronics, German Aerospace Center (DLR), Wessling, Germany

    Cosimo Della Santina

Authors
  1. Francesco Stella
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Cosimo Della Santina
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Josie Hughes
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Francesco Stella.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Machine Intelligence thanks David Howard for their contribution to the peer review of this work.

Supplementary information

Supplementary Information

Supplementary material

Rights and permissions

Reprints and Permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Stella, F., Della Santina, C. & Hughes, J. How can LLMs transform the robotic design process?. Nat Mach Intell 5, 561–564 (2023). https://doi.org/10.1038/s42256-023-00669-7

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s42256-023-00669-7

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing