EA ROBLOC

Associate Team ‘ROBLOC’: Robot Localization and Navigation

partners : INRIA Lyon Chroma and IntroLab / University of Sherbrooke / 3IT (Canada)

PIs : Olivier Simonin, Prof. (France) and François Michaud, Prof. (Canada)

The project started on March 2024.

YEAR 1 (2024)

Scientific results

In 2024 the ROBLOC associated team started with a collaborative work on simulation and control of fleet of drones (UAVs) to manage complex tasks requiring communication and reconfiguration, This work has been mainly developed during the internship of Théotime Balaguer (PhD student, Chroma) in Sherbrooke, from August to November, and during the internship of Etienne Villemure (Master student, IntroLab) in Lyon, from November to December (supervized by O. Simonin and F. Ferland).

This work and the first results can be summarized in several points :

  • T. Balaguer presented in Sherbrooke the DANCERS co-simulator he developed during his PhD to facilitate its use by Introlab colleagues, as it is an important tool in the EA ROBLOC (DANCERS allows to connect robotics simulators such as Gazebo with the NS3 network simulator).

  • T. Balaguer, E. Villemure and O. Simonin defined a new decentralized algorithm to manage the deployment of a fleet of drones from a source area to a target area. Our approach relies on the extension of the flocking model [Reynolds][Vicsek], which is initially defined to allow a fleet of drones to fly together. We defined an algorithm ensuring the emergence of a “mission” communication path from a leader drone to a source drone while the leader flies towards the target area. By defining two possible roles for others (mission node/free node), drones dynamically integrates the “mission” path while the flocking stretches until the leader robot reaches the target area.

  • To evaluate the efficiency and the limits of our algorithm, E. Villemure and T. Balaguer implemented the above presented algorithm and different test scenarios in DANCERS with Gazebo and NS3 (2D and 3D environments, w/wo obstacles, varying number of drones). They showed the algorithm is able to stretch the fleet (the flocking) while maintaining connectivity between the leader and the source drones. This work is in progress and will be continued in 2025.

Activity review

In 2024, we organized two missions :

  • a 3-month PhD internship in Sherbrooke/IntroLab by Théotime Balaguer (mid July – mid November 2024)
  • a 5 weeks internship in Lyon/Chroma team by Etienne Villemure, Master student (mid November – mid December 2024).

Production

  • Software developments are available on a new branch on the github Chroma-CITI/DANCERS repository.
  • Poster communication “UAV Relay Chain Emergence via Routing-aware Local Flocking Forces” T. Balaguer, E. Villemure, F. Ferland, I. Guérin Lassous, I. Fantoni, O. Simonin