-
Jeremy Fersula
- PHD
- Team: Amac
- Office: 303
- Email: fersula@isir.upmc.fr
- Website: https://www.linkedin.com/in/j%C3%A9r%C3%A9my-fersula-115014192/
- Bio: PhD student on the topic "Swarm Robotics: distributed online learning in the realm of active matter", the thesis is advised by t Nicolas Bredeche (ISIR, SU), Olivier Dauchot (Gulliver, ESPCI) and Marco Tarzia (LPTMC, SU). This experimental subject deals with swarms (~100) of robots that learn by embedding fully decentralized evolution algorithms. The aim is to: 1) study the collective dynamics of the robots from a physics point of view (active matter, complex systems, molecular dynamics simulation). 2) study the role of the morphology of robots on their ability to accomplish complex tasks. key-words : swarm robotics, active matter, distributed online learning
Publications
- Jeremy Fersula. Swarm Robotics : distributed Online Learning in the realm of Active Matter. Machine Learning [cs.LG]. Sorbonne Université, 2023. English. ⟨NNT : 2023SORUS494⟩. ⟨tel-04483298⟩
- Matan Yah Ben Zion, Jeremy Fersula, Nicolas Bredeche, Olivier Dauchot. Morphological computation and decentralized learning in a swarm of sterically interacting robots. Science Robotics, 2023, 8 (75), ⟨10.1126/scirobotics.abo6140⟩. ⟨hal-03840764⟩