I am currently involved in the following projects:
ANR VORTEX: Vision-based Reconfigurable Drone Swarms for Fast Exploration (2023-2027)
Coordinator: Olivier Simonin
My role: M2 intern adviser, co-adviser of Ph.D. student (since 2024)
Partners: CITI, ENSTA Paris, Lab-STICC, LS2N, ONERA
Description: The project proposes a new approach to exploring unknown indoor environments using a fleet of autonomous drones (UAVs). We propose to define a strategy based on swarm intelligence that exploits only vision-based behaviors. The fleet will deploy as a dynamic graph and self-reconfigure according to events and discovered areas. Without requiring mapping or wireless communication, the drones will coordinate by mutual perception and communicate by visual signs. This approach will be developed with RGB and event cameras to achieve fast and low-energy navigation. Performance, swarm properties, and robustness will be evaluated by building a demonstrator extending a quadrotor prototype developed by the consortium.
Former Projects
The projects I have been involved in the past:
ANR TRAITS: Trustworthy AI through Propagating Sets (2021 - 2024)
Coordinators: Goran Freshe and Matthias Althoff
My role: M1 intern adviser, co-adviser of Ph.D student (2022-2024)
Partners: ENSTA Paris (ENSTA), TU Munchen (TUM)
Description: The project focuses on developing a monitoring module to control systems using "black-box" algorithms. These algorithms are particularly suited to dynamic systems where unpredictable interactions make traditional modeling difficult, such as in autonomous vehicles or robotics. This project explores reinforcement learning (RL), an approach where agents learn based on their interactions with the environment, but which raises safety concerns. To address this challenge, a new method based on deterministic abstractions has been implemented, ensuring that agents comply with safety specifications while simplifying the synthesis of alternative controls if the agent's proposed control does not meet the specification. This approach thus enhances the reliability and safety of autonomous agents.
AID FARO: Le projet Fondements des Algorithmes de RObots en essaims (2021-2025)
Coordinator: Eric Goulbaut
My role: Researcher postdoctoral (2021-2022)
Partners: Ecole Polythehnique, ENSTA Paris, Telecom Paris, ENSTA Bretagne
Description: The FARO project focuses on the development of new algorithms (model-driven and/or data-driven) and new validation methods to ensure the proper functioning of drones in swarms, on land, in the air, or at sea.
ERC PROCSYS: Towards Programmable Cyber-physical Systems: a Symbolic Control Approach (2017-2023)
Coordinator: Antoine Girard
My role: PhD Student (2018-2021)
Partners: CNRS
Description: The PROCSYS project develops a high-level CPS language based on hybrid automata, enabling rich behavior specification and efficient controller synthesis. It addresses symbolic control challenges: scalability through parsimonious models and lazy synthesis and robustness via novel symbolic controller methods. The project advances distributed CPS contract design, data-driven abstraction, and symbolically guided predictive control.
