Carlo Bosio

Hi, I am Carlo! I am a PhD student at UC Berkeley advised by Mark W. Mueller within the Berkeley AI Research Lab. My research lies at the intersection of Robotics, Controls and Machine Learning.

Before coming to Berkeley, I obtained an MS with honors in Robotics from Sant'Anna School of Advanced Studies and the University of Pisa, Italy. I have also had the pleasure of spending time at EPFL (Switzerland), Stanford University (USA), and TBSI (China).

Feel free to reach out if you have any questions or would like to have a chat!

Email  |  CV  |  Google Scholar  |  LinkedIn

profile photo

News

  • [Oct 2024] Our preprint on Synthesizing Interpretable Control Policies with LLMs is available here!
  • [Sept 2024] I am an organizer of BARS 2025!
  • [Jun 2024] I am going to be part of the KTH RPL Summer School in Stockholm, Sweden!
  • [Apr 2024] Our work on ReachBot has been published on Science Robotics! Check it out here.
  • [Oct 2023] I am going to be at IROS 2023 in Detroit presenting this RA-L paper I worked on during my undergrad.
  • [Sep 2023] Our preprint on Robust Multi-Agent Aerial Manipulation is available here!
  • [Aug 2022] I started my PhD at Berkeley!

Research

I am interested in robotics, controls, multi-agent systems, robot learning. My current research focuses on developing algorithms to, by design, achieve robust behaviors in real world robotic systems. At the moment I am exploring the use of LLMs and evolutionary computation to help automate the design of robot morphologies and their controllers. Stay tuned!

project image

Synthesizing Interpretable Control Policies through Large Language Model Guided Search


Carlo Bosio, Mark W. Mueller
arXiv preprint arXiv:2410.05406, 2024
arxiv | video | abstract | bibtex

Encoding control policies as programs in Python and evolving them using LLMs.

project image

Locomotion as manipulation with ReachBot


Tony G. Chen, Stephanie Newdick, Julia Di, Carlo Bosio, Nitin Ongole, Mathieu LapĂ´tre, Marco Pavone, Mark R. Cutkosky
Science Robotics, 2024
arxiv | website | abstract | bibtex

A novel concept of rock climbing robot for Mars lava tube exploration.

project image

Automated Layout Design and Control of Robust Cooperative Grasped-Load Aerial Transportation Systems


Carlo Bosio, Jerry Tang, Ting-Hao Wang, Mark W. Mueller
arXiv preprint arXiv:2310.07649, 2023
arxiv | video | abstract | bibtex

Joint layout-control optimization for robustness of multi-drone aerial manipulation.





Design and source code can be found here.