As a technology transfer center, the CATIE organization has a broad set of skills and strong ties with engineer schools, laboratories and high tech startups. CATIE robotics was created in early 2018, the team aims to explore service robotics from an application-driven perspective and benefits from the multidisciplinary environment it lives in.
One of the team’s goals is to participate in the RoboCup @Home competition. The RoboCup competition is a way to objectively measure progress, expose it to a broad audience and gather an international community.
At a local scope, creating a robotics team is also a way of uniting people from various scientific horizons towards a very pragmatic, tangible goal.
Achievements, current state and near future
In early 2018, we chose the ROS based robot TIAGo from PAL Robotics as the main development platform. The robot was received in August 2018 and renamed Epock.
Since then, promising results have been achieved during 5 major robotics competitions: RoboCup @Home German Open 2019 (2nd place), RoboCup @Home Sidney 2019 (3rd place), SciRoc Challenge (2nd place), (virtual) RoboCup @Home 2021 (2nd place) and RoboCup @Home Bordeaux (3rd place)
A working pipeline of speech to text, Natural Language Understanding (NLU) and speech synthesis has been integrated and succesfully tested in the difficult environment of RoboCup challenges.
Furthermore, reliable safe navigation and accurate Simultaneous Localization And Mapping (SLAM) have been reached through the use of open source libraries.
State of the art neural networks architectures have been used for face recognition and body pose estimation. They have been integrated in the robot and showcased during the same competitions.
After having integrated previous developments to obtain robust and autonomous behaviors on the robot, the next step will be improving overall grasping capabilities and object recognition. We are also redesigning Epock’s head to enhance the interaction with users, particularly in order to make its intentions clearer to laymen.
Two projects related to object segmentation and grasping have been completed with the 2019 class Bordeaux INP students specializing in robotics as their main semester project. The 2020 class will work on flexible grasping.
Our (dream) team
Our (extremely supportive) extended team
Our (beloved) former members
Clément Duss’ (ieux)
« JB » Horel
Zhe « Aurore » Li
2021 – Virtual
SciRoc 2019 – Milton Keynes (United-Kingdom)
2023 – Bordeaux (France)
2019 – Sydney (Australia)
German Open 2019 – Magdeburg (Germany)
CATIE Robotics @Home 2024 Team Description Paper
by Sébastien DELPEUCH, Pierre-Marie ANCELE, Thierry ARSICAUD, Alban CHAUVEL, Charles DORMOY, Christine JAUREGUIBERRY, Clément LAIGLE, Clément PINET, Jean-Noël BARTHAS, Florian LARRUE and Sébastien LOTY
CATIE Robotics @Home 2023 Team Description Paper
by Sébastien DELPEUCH, Boris ALBAR, Alban CHAUVEL, Christine JAUREGUIBERRY, Clément LAIGLE, Sébastien GAMARDES, Clément PINET, Stéphane POUYET, Logan SAINT-GERMAIN, Florian LARRUE, Ludwig JOFFROY and Sébastien LOTY
CATIE Robotics @Home 2020 Team Description Paper
by Rémi FABRE, Boris ALBAR, Clément DUSSIEUX, Christine JAUREGUIBERRY, Ludwig JOFFROY, Jean-Baptiste HOREL, Zhe LI, Alexandre PERROT, Clément PINET, Stéphane POUYET, Florian LARRUE and Sébastien LOTY
CATIE Robotics @Home 2019 Team Description Paper
by Rémi FABRE, Boris ALBAR, Clément DUSSIEUX, Ludwig JOFFROY, Zhe LI, Clément PINET, Jennifer SIMEON and Sébastien LOTY
Methodology to Adapt Neural Network on Constrained Device at Topology level
by Logan SAINT-GERMAIN, Bertrand LE GAL, Fabien BALDACCI, Jeremie CRENNE, Christophe JEGO and Sébastien LOTY
Progress and first results in the study of the pilot-IA collaboration in SPO
by Charles-Alban DORMOY, Jean-Marc ANDRE, Théodore LETOUZE and Turkan HENTATI