Détail de l'offre

Master 2 Internship position on View Planning for 3D Reconstruction with Cooperative Multi-UAV Systems

Offre du 03/11/2022

A Master2 internship position is available at INRIA Lyon in the Chroma team (CITI-Lab) on the subject: "View Planning for 3D Reconstruction with Cooperative Multi-UAV Systems" under the supervision of Alessandro Renzaglia (alessandro.renzaglia@inria.fr)

Context. Autonomous exploration with a team of mobile robots is a fundamental task in many robotics applications, such as map reconstruction, search and rescue operations, inspections tasks, etc. For this reason, it is a widely studied problem and several efficient approaches have been proposed over the years. However, most of the existing solutions only consider 2D areas, while complex 3D environments still present several challenges to be addressed.

Objectives. The objective of this internship is to study existing approaches for autonomous 3D exploration and environment reconstruction with cooperating UAVs (unmanned aerial vehicles) and design new solutions to improve them. The main problem to solve will be the definition of next view points for the UAV fleet in order to maximize the future observations gain and so complete the mission in the minimum time possible. This will be based on two main tasks: the definition of criteria to quantify viewpoint utilities in terms of the predicted information gain, and the construction of a suitable set of viewpoint candidates to find the optimal one for each UAV. Standard next-best view algorithms usually exploit a-priori models of the 3D object to select the optimal viewpoints, while in an exploration scenario, probabilistic methods and information-based criteria need to be defined to obtain a prediction of the information gain. In a cooperative mission with multiple UAVs involved, the selection of viewpoints will also require taking into account possible overlaps and redundancies among different robots’ observations. Furthermore, the case of non-uniform information, where some areas need higher accuracy observations will be also considered. This study will be built upon the active research recently led in the Chroma team on the autonomous exploration of 3D environments and exploiting the Gazebo-based drones simulator developed by the team.

Experience required
Knowledge in robotics is required. Experience with multi-agent / multi-robot systems and/or planning algorithms will be appreciated.
Proficiency in Python or C++ programming is required. Experiences with ROS and Gazebo are valued.

Additional information
The selected candidate will join the INRIA Chroma team (https://team.inria.fr/chroma/) at CITI-lab INRIA/INSA Lyon, and work under the supervision of Alessandro Renzaglia. The internship duration is 5 or 6 months, ideally starting from February/March 2023.

To apply, send a CV and a motivation letter to:
Alessandro Renzaglia (alessandro.renzaglia@inria.fr)

CHROMA

38330 Montbonnot-Saint-Martin

alessandro.renzaglia@inria.fr

Type de Contrat

  • Stage
  • Fonction

    • Ingénieur.e
    • Diplômes

      • Bac +5 (M2, Ingénieur, etc.)
      • Expérience

        • Débutant
        • Compétences Techniques

          • Vision
          • Drone
          • Langages Informatiques

            • C, C++
            • Python
            • Outils et Environnements

              • ROS
              • Gazebo