Trustworthy AI for CCAM

Projektinformationen

AI4CCAM

ID Finanzhilfevereinbarung: 101076911

DOI

10.3030/101076911

Projekt abgeschlossen

EK-Unterschriftsdatum 21 Oktober 2022

Startdatum 1 Januar 2023

Enddatum 31 März 2026

Finanziert unter

• Climate, Energy and Mobility

Gesamtkosten

€ 5 965 630,00

EU-Beitrag

€ 5 965 630,00

Koordiniert durch

SIMULA RESEARCH LABORATORY AS
Norway

CORDIS bietet Links zu öffentlichen Ergebnissen und Veröffentlichungen von HORIZONT-Projekten.

Links zu Ergebnissen und Veröffentlichungen von RP7-Projekten sowie Links zu einigen Typen spezifischer Ergebnisse wie Datensätzen und Software werden dynamisch von OpenAIRE abgerufen.

Leistungen

Documents, reports (10)

AI4CCAM Validation Handbook

Use case implementation guidelines, KPI definition and validation methodologies. This deliverable is produced as part of T4.1.

Plan for dissemination and exploitation including communication activities (CDE)

Initial plan describing communication and dissemination activities. This deliverable is produced as part of T5.1

Integration of Human Gaze Models in Hybrid AI Models

Integrated report on the AI based scenario generation and prediction models integrating human gaze estimation models. This deliverable is produced as part of T2.2 and T2.3.

Innovation Registry and Patent Filing

Elicitation of project innovations in AI4CCAM registry and two patents filed. This deliverable is produced as part of T5.3

Plan for dissemination and exploitation including communication activities (CDE) - Initial Report and Updated Plan

Initial report on communication and dissemination activities and updated plan. This deliverable is produced as part of T5.1

AI4CCAM Use Cases Implementation and Validation

Report including KPI measurements, validation criteria and results, impact assessment, best practises and lessons learnt. This deliverable is produced as part of T4.2, T4.3, T4.4 and T4.5

AI4CCAM Trustworty AI Documentation Framework - Final Version

Inclusion of AI risks and cyber-threats, stakeholders' perceptions and biases, end-users and VRUs perceptions with recommendations on user acceptance. This deliverable is produced as part of T3.2.

AI4CCAM Trustworthy AI Documentation Framework - Initial Version

Guidelines for fairness and diversity, for improved transparency-led user agency, communication, and acceptance; Governance techniques for accountability in AI-based CCAM operations with ethical dilemma identification. This deliverable is produced as part of T3.2.

Methodology for Trustworthy AI in CCAM

Methodology including scenario description (9) with functional and Trustworthy AI requirements considering AI standards. This deliverable is produced as part of T1.1, T1.2 and T1.3

Plan for dissemination and exploitation including communication activities (CDE) - Final Report

Final report on communication and dissemination activities. This deliverable is produced as part of T5.1

Data Management Plan (1)

Data Management Plan (DMP) - Final Report

Final report on data management in the project. This deliverable is produced as part of Task 6.3.

Other (3)

Data-driven simulation scenarios

Scenarios containing use cases simulation of VRUs interactions with CAVs. This deliverable is produced as part of T4.3

Participatory AI4CCAM Space - Final Version

Final version with AI risks and security threats discussion threads, data bias identification list, end-users and VRUs perception topics. This deliverable is produced as part of T3.3.

Participatory AI4CCAM Space - Initial Version

Initial version of the participatory space with Trustworthy AI in CCAM, CAV User Acceptance topics and Q/A. This deliverable is produced as part of T3.1

Veröffentlichungen

Peer reviewed articles (3)

Towards Trustworthy Automated Driving through Qualitative Scene Understanding and Explanations

Autoren: Nassim Belmecheri, Arnaud Gotlieb, Nadjib Lazaar and Helge Spieker
Veröffentlicht in: SAE International Journal of Connected and Automated Vehicles, 2024, ISSN 2574-0741
Herausgeber: SAE International
DOI: 10.48550/arXiv.2403.16908

Query-driven Qualitative Constraint Acquisition

Autoren: Mohamed-Bachir Belaid, Nassim Belmecheri, Arnaud Gotlieb, Nadjib Lazaar, Helge Spieker
Veröffentlicht in: Journal of Artificial Intelligence Research, Ausgabe 79, 2024, ISSN 1076-9757
Herausgeber: AAAI Press
DOI: 10.1613/jair.1.14752

Boosting the learning for ranking patterns

Autoren: Nassim Belmecheri, Noureddine Aribi, Nadjib Lazaar, Yahia Lebbah, Samir Loudni
Veröffentlicht in: algorithms, 2023, ISSN 1999-4893
Herausgeber: MDPI Open Access Publishing
DOI: 10.48550/arXiv.2203.02696

Other (3)

Acquiring Qualitative Explainable Graphs for Automated Driving Scene Interpretation

Autoren: Belmecheri, Nassim; Gotlieb, Arnaud; Lazaar, Nadjib; Spieker, Helge
Veröffentlicht in: arXiv, 2023
Herausgeber: Arxiv
DOI: 10.48550/arxiv.2308.12755

Trustworthy Automated Driving through Qualitative Scene Understanding and Explanations

Autoren: Nassim Belmecheri, Arnaud Gotlieb, Nadjib Lazaar, Helge Spieker
Veröffentlicht in: arXiv, 2024
Herausgeber: arXiv
DOI: 10.48550/arXiv.2403.09668

GSGFormer: Generative Social Graph Transformer for Multimodal Pedestrian Trajectory Prediction

Autoren: Zhongchang Luo, Marion Robin and Pavan Vasishta
Veröffentlicht in: arXiv, 2023
Herausgeber: arXiv
DOI: 10.48550/arXiv.2312.04479

Conference proceedings (1)

Evaluating Human Trajectory Prediction with Metamorphic Testing

Autoren: Helge Spieker, Nassim Belmecheri, Arnaud Gotlieb, Nadjib Lazaar
Veröffentlicht in: MET'24: 9th ACM International Workshop on Metamorphic Testing, 2024
Herausgeber: ISSTA/ECOOP 2024
DOI: 10.48550/arXiv.2407.18756