@ CNRS-I3S/UNS

Research carried out by our members is focused on  :

  • Real-time visual localisation : robustness, accuracy, efficiency.
  • Robot learning : Dense 3D mapping, semantic segmentation, visual SLAM, large-scale scenes, dynamic environments, loop-closure.
  • Sensor models : RGB-D, camera, stereo, depth, omni-directional, laser.
  • Applications to Robotics and Augmented Reality.
  • Postdoc Position Filled in the context of the European H2020 projet Comanoid (CNRS, INRIA, DLR, Univ. Rome 1, Airbus) on RGB-D localisation and mapping for humanoid robots in a manufacturing environment - more information is available here.
  • PhD defence of Fernando Ireta on the 4th of April 2018 on the topic of "Estimation de pose globale et suivi pour la localisation RGB-D et cartographie 3D". The PhD was supervised by Andrew Comport and the jury was composed of Pascal Vasseur (Université de Rouen), Youcef Mezouar (Institute Pascal), Jonghyuk Kim, (Australian National University), Christian Laugier (INRIA) and Jean-Bernard Hayet, (CIMAT, Mexico).
  • Robot PACA, steering committee member of RobotPACA − Workshop to unite all the actors in robotics throughout the region PACA, 25-26 juin 2018, Sophia Antipolis, France.
  • IIT Visual Geometry and Modelling group, invited talk, Andrew Comport, Dense visual SLAM for autonomous navigation and augmented reality, 19th-20th of June, 2017.
  • Arnaud Tanguy, Pierre Gergondet, Andrew Comport, Abderrahmane Kheddar, Award of the best paper finalist for the article IEEE/SICE SII entitled “Closed-loop RGB-D SLAM Multi-Contact Control for Humanoid Robots” presented at SII 2016 (International Symposium on System Integration)

  • Fernando Ireta and Andrew Comport, Award of the best student paper finalist at IEEE/RSJ IROS, for the article entitled “Point-to-hyperplane RGB-D Pose Estimation : Fusing Photometric and Geometric Measurements”. Prestented at IROS 2016.
  • Fernando Ireta and Andrew Comport, Awarded the best student paper finalist at IEEE MFI, for the article entitled “A Proof that Fusing Measurements Using Point-to-Hyperplace Registration is Invariant to Relative Scale”. Presented at "Multi-sensor Fusion and Integration for Intelligent Systems".
  • PIXMAP, Prize-winner for the French national innovation competition CREADEV for the creation of start-ups.
  • PhD position filled on Object recognition in RGB-D images For the accurate localisation of a manufacturing collaborative robot in collaboration with Airbus Group.
  • Andrew Comport and Maxime Meilland, Awarded the medal of the University of Nice Sophia Antipolis 2014.
  • European H2020 projet Comanoid on Multi-contact Collaborative Humanoids in Aircraft Manufacturing begins in collaboration with CNRS, INRIA, Airbus, DLR et l'Université de Rome 1.
  • We are looking for a postdoc to work on large-scale dense visual SLAM. More information is available HERE. Please contact us rapidly if you are interested.
  • Invited talk at the Robotics and Perception Group at the University of Zurich.
  • Invited conference on the "Next generation of collaborative robots" at the Embassy of France in Japan sponsored by Airbus and the French embassy, 26/2/2014.
  • IEEE/RSJ IROS Best Paper Award : On unifying key-frame and voxel-based dense visual SLAM at large scales, 2013, Maxime Meilland and Andrew Comport

airbusnvidia vector logo        H2020c                                     CNRSfilaire grand           logo400x160 99dpi      NewlogoI3S