Robotic Safe Adaptation In unprecedented Situations

Informacje na temat projektu

RoboSAPIENS

Identyfikator umowy o grant: 101133807

DOI

10.3030/101133807

Data podpisania przez KE 14 Listopada 2023

Data rozpoczęcia 1 Stycznia 2024

Data zakończenia 31 Grudnia 2026

Finansowanie w ramach

• Digital, Industry and Space

Koszt całkowity

€ 6 883 242,50

Wkład UE

€ 6 883 233,50

Koordynowany przez

AARHUS UNIVERSITET
Denmark

CORDIS oferuje możliwość skorzystania z odnośników do publicznie dostępnych publikacji i rezultatów projektów realizowanych w ramach programów ramowych HORYZONT.

Odnośniki do rezultatów i publikacji związanych z poszczególnymi projektami 7PR, a także odnośniki do niektórych konkretnych kategorii wyników, takich jak zbiory danych i oprogramowanie, są dynamicznie pobierane z systemu OpenAIRE .

Rezultaty

Demonstrators, pilots, prototypes (1)

Architecture for the RoboSAPIENS platform Loop with safety and trustworthiness

First version of the deployment on platform from D5.1, with guidelines and lab-scale validation plans.

Documents, reports (4)

Modeling and simulation

Modeling and simulation of the case studies, validated to ensure changes can be made during simulation.

Requirements for safe and trustworthy MAPE-K loops

Ontology for domain concepts and requirements of MAPE-K, including feature models for the domain of safety and use cases.

Formal models including uncertainty quantification and patterns for the PAL and DTI architectures

Initial version of the architectural pattern, models, and uncertainty quantification methods, for the PAL and DTI architectures, and their formal semantics.

Case study compendium

Specifications of the case studies, investigation of baseline technologies, and assessment criteria to be used in the case studies including the people’s uncertainties.

Other (5)

Active learning and system monitoring

Initial (a) uncertainty estimator for the ship predictor and method for detecting inability for disassembling tasks; (b) methods for enriching scheduling algorithms; and (c) active perception technology.

Operational requirements for the case studies

Initial version of the controlled natural language and operational requirements for case studies, and their semantics.

Initial architecture for the RoboSAPIENS platform

Hardware and software setup for a non-distributed platform, and workflow fordeployment of the initial architecture on it.

Monitorable and trustworthy verification loops

Language to describe monitorable, safe, and trustworthy MAPE-K loops: grammar and tool.

Self-adaptation and initial safety integration

Initial version of (a) functional and safety requirements for use cases; (b) tools forpersistently exciting systems, and active perception. Detection methods for environmental and structural changes.

Publikacje

Conference proceedings (6)

Improving Active Face Recognition using Multi-Task Learning

Autorzy: Michail Loufakis, Pavlos Tosidis, Nikolaos Passalis, Anastasios Tefas
Opublikowane w: 15th International Conference on Information, Intelligence, Systems and Applications (IISA) 2024, 2024
Wydawca: IEEE

Improving Active Face Recognition using Multi-Task Learning

Autorzy: Michail Loufakis, Pavlos Tosidis, Nikolaos Passalis, Anastasios Tefas
Opublikowane w: 2024 15th International Conference on Information, Intelligence, Systems & Applications (IISA), 2024
Wydawca: IEEE
DOI: 10.1109/IISA62523.2024.10786720

Assessing the Uncertainty and Robustness of the Laptop Refurbishing Software

Autorzy: Chengjie Lu, Jiahui Wu, Shaukat Ali, Mikkel Labori Olsen
Opublikowane w: 2025 IEEE Conference on Software Testing, Verification and Validation (ICST), 2025
Wydawca: IEEE
DOI: 10.1109/ICST62969.2025.10988977

Continual Updates of Face Embeddings using Deep Reinforcement Learning

Autorzy: Pavlos Tosidis, Nikolaos Passalis and Anastasios Tefas
Opublikowane w: 27th International Conference on Pattern Recognition (ICPR) 2024, Workshop on Facial and Body Expressions, 2024
Wydawca: IAPR

Runtime Verification of Autonomous Systems Utilizing Digital Twins as a Service

Autorzy: Morten Haahr Kristensen, Alberto Bonizzi, Cláudio Gomes, Simon Thrane Hansen, Carlos Isasa, Hannes Iven, Eduard Kamburjan, Peter Gorm Larsen, Martin Leucker, Prasad Talasila, Valdemar Trøjgård Tang, Stefano Tonetta, Lars B. Vosteen, Thomas Wright
Opublikowane w: 2024 IEEE International Conference on Autonomic Computing and Self-Organizing Systems Companion (ACSOS-C), 2024
Wydawca: IEEE
DOI: 10.1109/ACSOS-C63493.2024.00042

Probabilistic Knowledge Transfer for Lightweight Vision Transformers

Autorzy: Ioannis Bountouris, Nikolaos Passalis, Anastasios Tefas
Opublikowane w: 32nd European Signal Processing Conference (EUSIPCO), 2024, 2024
Wydawca: EURASIP

Book chapters (4)

Formal Architectural Patterns for Adaptive Robotic Software

Autorzy: James Baxter, Bert van Acker, Morten Kristensen, Thomas Wright, Ana Cavalcanti, Cláudio Gomes
Opublikowane w: Lecture Notes in Computer Science, Fundamental Approaches to Software Engineering, 2025
Wydawca: Springer Nature Switzerland
DOI: 10.1007/978-3-031-90900-9_8

Foundation Models for the Digital Twins Creation of Cyber-Physical Systems

Autorzy: Shaukat Ali, Paolo Arcaini, Aitor Arrieta
Opublikowane w: Lecture Notes in Computer Science, Leveraging Applications of Formal Methods, Verification and Validation. Application Areas, 2024
Wydawca: Springer Nature Switzerland
DOI: 10.1007/978-3-031-75390-9_2

Process-Algebraic Semantics for Verifying Intelligent Robotic Control Software

Autorzy: Ziggy Attala, Fang Yan, Simon Foster, Ana Cavalcanti, Jim Woodcock
Opublikowane w: Lecture Notes in Computer Science, NASA Formal Methods, 2025
Wydawca: Springer Nature Switzerland
DOI: 10.1007/978-3-031-93706-4_2

Large Models in Dialogue for Active Perception and Anomaly Detection

Autorzy: Tzoulio Chamiti, Nikolaos Passalis, Anastasios Tefas
Opublikowane w: Lecture Notes in Computer Science, Pattern Recognition, 2024
Wydawca: Springer Nature Switzerland
DOI: 10.1007/978-3-031-78447-7_25

Peer reviewed articles (4)

Robotic safe adaptation in unprecedented situations: the RoboSAPIENS project

Autorzy: Larsen, G. Peter; Ali, Shaukat; Behrens, Roland; Cavalcanti, Ana; Gomes, Claudio; Li, Guoyuan; Meulenaere, Paul De; Olsen, Mikkel L.; Passalis, Nikolaos; Peyrucain, Thomas; Tapia, Jesús; Tefas, Anastasios; Zhang, Houxiang
Opublikowane w: Research Directions: Cyber-Physical Systems, 2024, ISSN 2753-8516
Wydawca: Cambridge University Press
DOI: 10.1017/CBP.2024.4

Learning Nonlinear Dynamics of Ocean Surface Vessel With Multistep Constraints

Autorzy: Tongtong Wang, Robert Skulstad, Motoyasu Kanazawa, Guoyuan Li, Houxiang Zhang
Opublikowane w: IEEE Transactions on Industrial Informatics, Numer 20, 2024, ISSN 1551-3203
Wydawca: Institute of Electrical and Electronics Engineers (IEEE)
DOI: 10.1109/TII.2024.3396359

Physics-Informed Neural Networks for Robust System Identification of Ship Roll Dynamics With Noise Resilience

Autorzy: Tongtong Wang, Robert Skulstad, Houxiang Zhang
Opublikowane w: IEEE Transactions on Industrial Informatics, Numer 21, 2025, ISSN 1551-3203
Wydawca: Institute of Electrical and Electronics Engineers (IEEE)
DOI: 10.1109/TII.2025.3534406

Leveraging active perception for real-time high-resolution pose estimation

Autorzy: Theodoros Manousis, Emmanouil Eleftheriadis, Nikolaos Passalis, Anastasios Tefas
Opublikowane w: Expert Systems with Applications, Numer 271, 2025, ISSN 0957-4174
Wydawca: Elsevier BV
DOI: 10.1016/J.ESWA.2025.126550