Next generation of AI-powered robotics for agile production

Informations projet

AGIMUS

N° de convention de subvention: 101070165

DOI

10.3030/101070165

Date de signature de la CE 20 Juin 2022

Date de début 1 Octobre 2022

Date de fin 30 Septembre 2026

Financé au titre de

• Digital, Industry and Space

Coût total

€ 5 278 250,00

Contribution de l’UE

€ 4 999 187,50

Coordonné par

CENTRE NATIONAL DE LA RECHERCHE SCIENTIFIQUE CNRS
France

CORDIS fournit des liens vers les livrables publics et les publications des projets HORIZON.

Les liens vers les livrables et les publications des projets du 7e PC, ainsi que les liens vers certains types de résultats spécifiques tels que les jeux de données et les logiciels, sont récupérés dynamiquement sur OpenAIRE .

Livrables

Other (6)

Sensor-driven MPC with visual-haptic feedback - First Version

First version of the sensor-driven MPC controller with visual-haptic feedback, as developed through T4.3

Efficient and versatile differentiable simulator - First Version

The first version of the efficient and versatile differentiable physics simulation software developed through T2.1

Motion dataset from the planner created through supervised learning

A first version of a motion dataset module (T3.2), that memorizes trajectories from the task and motion planner (T3.4) and is able to generate a set of various trajectories that are automatically adapted to a given scenario at the inference time.

6D object pose estimation

A high-accuracy and robust vision perception module for estimating object 6D pose, as developed through T4.1

Trajectory optimizer with contact invariance

The first version of the trajectory optimisation software developed able to handle hard equality and inequality constraints (T2.2) by incorporating also contact invariance techniques (T2.3)

Planner guided by demonstration extracted from a monocular camera - First Version

First version of the planning software guided by demonstration extracted from a monocular camera developed through T3.1

Data Management Plan (2)

Data Management Plan - Second Version

The interim updated version of the Data Management Plan. This version will build upon and update D8.2 to reflect any newly identified data that were collected / generated during the project as well as any changes in data previously identified and/or the methodology and plans of AGIMUS for sound and FAIR data management.

Data Management Plan - First Version

A report produced under T8.3. The report will outline the data to be collected / generated during the project as well as the data management methodology and plan to be followed across their life cycle in order to ensure their sound and FAIR management in line with applicable legislation at national and EU level.

Documents, reports (5)

Specification of the industrial pilot case studies

A report consolidating the specification of the three AGIMUS pilot case studies under T6.1.

Agile Production Deep Dive Analysis for next generation robotics by design

A report outlining the findings of the mapping exercise and the respective analysis conducted during T1.1 along with the methodology employed to produce them, along with the ethical, trustworthy, and secure by-design principles extracted during T1.2, including end-user and Advisory Board engagement activities and their results.

Dissemination and Communication Plan and Activities - Second Version

The updated version of D7.1 fine-tuning as appropriate the strategy, plan and actions to be undertaken under T7.1 and T7.4 with respect to dissemination, communication, synergies, and knowledge sharing activities of the project. The report will also update, if necessary, the targets to be met by the respective activities of the project throughout its course.

Specification of the AGIMUS Testing Zones, Robots and Skills

A report elaborating on the specifications of the three AGIMUS testing zones, the requirements and upgrades for the available and developed robotic systems, as well as the specification and design of the use cases for the deploying and testing the necessary robotic skills, as defined through T5.1 and T5.2

Dissemination and Communication Plan and Activities - First Version

A report elaborating on the strategy, plan, and actions to be undertaken under T7.1 and T7.4 with respect to dissemination, communication, synergies, and knowledge sharing activities of the project. The report will also include specific qualitative and quantitative targets to be met by the respective activities of the project throughout its course.

Publications

Conference proceedings (10)

Model Predictive Control Under Hard Collision Avoidance Constraints for a Robotic Arm

Auteurs: Haffemayer, Arthur; Jordana, Armand; Fourmy, Médéric; Wojciechowski, Krzysztof; Saurel, Guilhem; Petrík, Vladimír; Lamiraux, Florent; Mansard, Nicolas
Publié dans: 2024 21st International Conference on Ubiquitous Robots (UR), 2024
Éditeur: IEEE
DOI: 10.1109/UR61395.2024.10597485

Enforcing the consensus between Trajectory Optimization and Policy Learning for precise robot control

Auteurs: Quentin Le Lidec, Wilson Jallet, Ivan Laptev, Cordelia Schmid, Justin Carpentier
Publié dans: IEEE International Conference on Robotics and Automation (ICRA), 2023, ISSN 2331-8422
Éditeur: arXiv
DOI: 10.48550/arXiv.2209.09006Focustolearnmore

MegaPose: 6D Pose Estimation of Novel Objects via Render & Compare

Auteurs: Labbé, Yann; Manuelli, Lucas; Mousavian, Arsalan; Tyree, Stephen; Birchfield, Stan; Tremblay, Jonathan; Carpentier, Justin; Aubry, Mathieu; Fox, Dieter; Sivic, Josef
Publié dans: CoRL 2022 - Conference on Robot Learning, Dec 2022, Auckland, New Zealand, 2022
Éditeur: CORL
DOI: 10.48550/arXiv.2212.06870

From Compliant to Rigid Contact Simulation: a Unified and Efficient Approach

Auteurs: Carpentier, Justin; Le Lidec, Quentin; Montaut, Louis
Publié dans: Robotics: Science and Systems XX, 2024
DOI: 10.48550/ARXIV.2405.17020

Force Feedback Model-Predictive Control via Online Estimation

Auteurs: Jordana, Armand; Kleff, Sébastien; Carpentier, Justin; Mansard, Nicolas; Righetti, Ludovic
Publié dans: 2024 IEEE International Conference on Robotics and Automation (ICRA), 2024
DOI: 10.1109/ICRA57147.2024.10611156

FIGAROH: A Python Toolbox for Dynamic Identification and Geometric Calibration of Robots and Humans

Auteurs: Thanh D. V. Nguyen; Vincent Bonnet; Maxime Sabbah; Maxime Gautier; Pierre Fernbach; Florent Lamiraux
Publié dans: 2023
Éditeur: IEEE
DOI: 10.1109/HUMANOIDS57100.2023.10375232

Differentiable Collision Detection: a Randomized Smoothing Approach

Auteurs: Louis Montaut, Quentin Le Lidec, Antoine Bambade, Vladimir Petrik, Josef Sivic, Justin Carpentier
Publié dans: IEEE International Conference on Robotics and Automation (ICRA), 2023, ISBN 979-8-3503-2365-8
Éditeur: IEEE
DOI: 10.48550/arXiv.2209.09012

Multi-Contact Task and Motion Planning Guided by Video Demonstration

Auteurs: Kateryna Zorina, David Kovar, Florent Lamiraux, Nicolas Mansard (2), Justin Carpentier, Josef Sivic, Vladimir Petrik
Publié dans: IEEE International Conference on Robotics and Automation (ICRA), 2023, ISBN 979-8-3503-2365-8
Éditeur: IEEE
DOI: 10.1109/ICRA48891.2023.10161551

Towards a new generation of TIAGo robots

Auteurs: Ferro, Francesco; Marchionni, Luca; Banos, Narcis Miguel
Publié dans: 2023
Éditeur: Fifth Italian Conference on Robotics and Intelligent Machines
DOI: 10.5281/ZENODO.10722596

Linear-time Differential Inverse Kinematics: an Augmented Lagrangian Perspective

Auteurs: Wingo, Bruce; Sathya, Ajay; Caron, Stéphane; Hutchinson, Seth; Carpentier, Justin
Publié dans: Robotics: Science and Systems XX, 2024
DOI: 10.15607/RSS.2024.XX.110

Peer reviewed articles (4)

Contact Models in Robotics: a Comparative Analysis

Auteurs: Quentin Le Lidec, Wilson Jallet, Louis Montaut, Ivan Laptev, Cordelia Schmid, Justin Carpentier
Publié dans: IEEE Transactions on Robotics, 2023, ISSN 2331-8422
Éditeur: IEEE
DOI: 10.48550/ARXIV.2304.06372

Imitrob: Imitation Learning Dataset for Training and Evaluating 6D Object Pose Estimators

Auteurs: Jiri Sedlar, Karla Stepanova, Radoslav Skoviera, Jan K. Behrens, Matus Tuna, Gabriela Sejnova, Josef Sivic, Robert Babuska
Publié dans: IEEE Robotics and Automation Letters, 2023, ISSN 2377-3766
Éditeur: IEEE
DOI: 10.1109/LRA.2023.3259735

GJK++: Leveraging Acceleration Methods for Faster Collision Detection

Auteurs: Louis Montaut, Quentin Le Lidec, Vladimír Petrík, Josef Sivic, Justin Carpentier
Publié dans: IEEE Transactions on Robotics, 2024, ISSN 1552-3098
Éditeur: Institute of Electrical and Electronics Engineers
DOI: 10.1109/TRO.2024.3386370

Temporally Consistent Object 6D Pose Estimation for Robot Control

Auteurs: Kateryna Zorina, Vojtech Priban, Mederic Fourmy, Josef Sivic, Vladimir Petrik
Publié dans: IEEE Robotics and Automation Letters, Numéro 10, 2024, ISSN 2377-3766
Éditeur: Institute of Electrical and Electronics Engineers (IEEE)
DOI: 10.1109/LRA.2024.3502052

Other (9)

QPLayer: efficient differentiation of convex quadratic optimization

Auteurs: Antoine Bambade, Fabian Schramm, Adrian Taylor, Justin Carpentier
Publié dans: 2023, ISSN 2267-828X
Éditeur: HAL

Reconciling RaiSim With the Maximum Dissipation Principle

Auteurs: Quentin Le Lidec; Justin Carpentier
Publié dans: IEEE Transactions on Robotics, 2024
Éditeur: IEEE
DOI: 10.1109/TRO.2024.3430711

Model predictive control under hard collision avoidance constraint for a robotic arm

Auteurs: Haffemayer, Arthur; Jordana, Armand; Fourmy, Médéric; Wojciechowski, Krzysztof; Lamiraux, Florent; Mansard, Nicolas
Publié dans: 2024
Éditeur: HAL

End-to-End and Highly-Efficient Differentiable Simulation for Robotics

Auteurs: Le Lidec, Quentin; Montaut, Louis; de Mont-Marin, Yann; Carpentier, Justin
Publié dans: 2024
Éditeur: HAL
DOI: 10.48550/ARXIV.2409.07107

PROXQP: an Efficient and Versatile Quadratic Programming Solver for Real-Time Robotics Applications and Beyond

Auteurs: Bambade, Antoine; Schramm, Fabian; Kazdadi, Sarah, El; Caron, Stéphane; Taylor, Adrien; Carpentier, Justin
Publié dans: IEEE Transactions on Robotics, 2023
Éditeur: IEEE

Constrained Articulated Body Algorithms for Closed-Loop Mechanisms

Auteurs: Ajay Suresha Sathya; Justin Carpentier
Publié dans: 2025
Éditeur: HAL

Visually Guided Model Predictive Robot Control via 6D Object Pose Localization and Tracking

Auteurs: Mederic Fourmy, Vojtech Priban, Jan Kristof Behrens, Nicolas Mansard, Josef Sivic, Vladimir Petrik
Publié dans: 2023, ISSN 2331-8422
Éditeur: arXiv
DOI: 10.48550/arXiv.2311.05344

Constrained Articulated Body Dynamics Algorithms

Auteurs: Ajay Suresha Sathya; Justin Carpentier
Publié dans: IEEE Transactions on Robotics, 2025
Éditeur: IEEE
DOI: 10.1109/TRO.2024.3502515

Risk-Sensitive Extended Kalman Filter

Auteurs: Jordana, Armand; Meduri, Avadesh; Arlaud, Etienne; Carpentier, Justin; Righetti, Ludovic
Publié dans: 2024 IEEE International Conference on Robotics and Automation (ICRA), 2024
Éditeur: IEEE
DOI: 10.48550/ARXIV.2305.11573