AI Powered human-centred Robot Interactions for Smart Manufacturing

Informazioni relative al progetto

AI-PRISM

ID dell’accordo di sovvenzione: 101058589

DOI

10.3030/101058589

Progetto chiuso

Data della firma CE 24 Maggio 2022

Data di avvio 1 Ottobre 2022

Data di completamento 30 Settembre 2025

Finanziato da

• Digital, Industry and Space

Costo totale

€ 10 827 138,76

Contributo UE

€ 9 335 578,88

Coordinato da

NTT DATA SPAIN, SL
Spain

CORDIS fornisce collegamenti ai risultati finali pubblici e alle pubblicazioni dei progetti ORIZZONTE.

I link ai risultati e alle pubblicazioni dei progetti del 7° PQ, così come i link ad alcuni tipi di risultati specifici come dataset e software, sono recuperati dinamicamente da .OpenAIRE .

Risultati finali

Documents, reports (4)

Task Analysis for benchmarking and design requirements (I)

Written report providing in-depth analysis of the original manual tasks for each use case, setting out risks and recommended mitigations, guidance for function allocation, and human-centred design requirements. Behavioural models and skills analysis will also be included in some of the iterations (M20 and M36 respectively).

AI-PRISM Dissemination and Communication Report (I)

Documentation of the actions and outcomes achieved concerning the promotional strategies. The final report will detail the international cooperation report.

Technology benchmarking and Project Vision

This deliverable will provide details about project vision and the benchmarking oftechnology landscape. It will be the main output of tasks 1.1 and 1.2.

AI-PRISM Impact Master Plan

Project’s dissemination, exploitation, and communication plans, outlining the target groups and segments. The report will include a description of the project website and social media, and the standardisation landscape.

Other (2)

AI-PRISM Human Robot Cooperation Ambient (I)

Consecutive releases of the CI/CD framework (code repositories, CI/CD pipelines and project scaffolds to support the development, integration, and validation of AI-based modules). Software releases of libraries of AI-based modules to enhance perception andreasoning capabilities of agents in the collaborative ambient, as well as to enable ambient level coordination and learning tasks by human demonstration.

AI-PRISM Human Centred Collaborative Robotic Platform (I)

Consecutive releases of the AI-PRISM software which includes libraries for real time communication, ambient digitalisation, and simulation. This includes code repositories, and pipelines to support the development of new modules, their deployment and testing. The simulation platform supports specific developed modules for agent simulation and for collaboration and risks evaluation.

Pubblicazioni

Conference proceedings (7)

A Study to Compare YOLO Family Models for Metal Surface Anomaly Detection

Autori: A. Koc, O. Albayrak, Y. Kurt and P. Unal
Pubblicato in: IEEE International Conference on Big Data (BigData), 2023
Editore: IEEE
DOI: 10.1109/BigData59044.2023.10386849

Revisiting Fully Convolutional Geometric Features for Object 6D Pose Estimation

Autori: Corsetti, Jaime; Boscaini, Davide; Poiesi, Fabio
Pubblicato in: 8th International Workshop on Recovering 6D Object Pose, 2023
Editore: 2023 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW)
DOI: 10.48550/arxiv.2307.15514

Breaking the barriers: multilingual user engagement to increase process engagement and technology acceptance in manufacturing

Autori: Iveta Eimontaite, Sarah Fletcher, Krystian Goławski, Tomasz Kołcon
Pubblicato in: Human Aspects of Advanced Manufacturing, Numero Volume 80, 2023
Editore: AHFE International
DOI: 10.54941/AHFE1003503

FreeZe: Training-free zero-shot 6D pose estimation with geometric and vision foundation models

Autori: Caraffa, Andrea; Boscaini, Davide; Hamza, Amir; Poiesi, Fabio
Pubblicato in: Computer Vision and Pattern Recognition, 2024
Editore: arXiv

Novel Class Discovery for 3D Point Cloud Semantic Segmentation

Autori: Riz, Luigi; Saltori, Cristiano; Ricci, Elisa; Poiesi, Fabio
Pubblicato in: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023
Editore: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
DOI: 10.48550/arxiv.2303.11610

TO COLLABORATE OR NOT TO COLLABORATE? HOW TO DETERMINE THE MOST SUITABLE LEVEL OF AUTOMATION TO INCREASE WORKFORCE SUSTAINABILITY AND PRODUCTION EFFICIENCY

Autori: SACHA GODHANIA, IVETA EIMONTAITE, SARAH R. FLETCHER, ANA GONZALEZ SEGURA, ELOY RUIZ, ANA PAOLA CARO OSPINA
Pubblicato in: ICRES 2023: 8th International Conference on Robot Ethics and Standards, 2023
Editore: CLAWAR Association - Series on Robot Ethics and Standards
DOI: 10.13180/icres.2023.17-18.07.006

Open-vocabulary object 6D pose estimation

Autori: Jaime Corsetti, Davide Boscaini, Changjae Oh, Andrea Cavallaro, Fabio Poiesi
Pubblicato in: Computer Vision and Pattern Recognition, 2024
Editore: Computer Vision and Pattern Recognition
DOI: 10.48550/arXiv.2312.00690

Peer reviewed articles (3)

Adaptive Beacon Period Configurator for Scalable LoRaWAN Downlink Applications

Autori: SILVESTRE-BLANES, Javier
Pubblicato in: IEEE Access, 2023, ISSN 2169-3536
Editore: IEEE Access
DOI: 10.1109/ACCESS.2023.3301991

PatchMixer: Rethinking network design to boost generalization for 3D point cloud understanding

Autori: Davide Boscaini; Fabio Poiesia
Pubblicato in: Image and Vision Computing, 2023, ISSN 0262-8856
Editore: Image and Vision Computing
DOI: 10.2139/ssrn.4370105

Novel class discovery meets foundation models for 3D semantic segmentation

Autori: Riz, Luigi; Saltori, Cristiano; Wang, Yiming; Ricci, Elisa; Poiesi, Fabio
Pubblicato in: International Journal of Computer Vision, 2024, ISSN 1573-1405
Editore: International Journal of Computer Vision
DOI: 10.48550/arxiv.2312.03782