Advanced integrated obstacle and track intrusion detection system for smart automation of rail transport

Informations projet

SMART2

N° de convention de subvention: 881784

DOI

10.3030/881784

Projet clôturé

Date de signature de la CE 29 Novembre 2019

Date de début 1 Decembre 2019

Date de fin 30 Novembre 2022

Financé au titre de

• SOCIETAL CHALLENGES - Smart, Green And Integrated Transport

Coût total

€ 1 708 737,50

Contribution de l’UE

€ 1 499 528,75

Coordonné par

UNIVERSITAET BREMEN
Germany

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

Documents, reports (10)

Report on sub-systems conformance testing

Deliverable reports on the outcome of Task 51 on conformance testing of all the subsystems to be integrated into a functional holistic ODTID SMART2 system

Algorithms for decision making

Deliverable reports on the algorithms developed in Task 41 for the fusion of preprocessed data of onboard and TSairborne ODTID subsystems

Recommendation for new set of lower complexity requirements for obstacle and track intrusion detection system

Deliverable report presenting the outcomes of Task 15 assessment of requirements for obstacle detection and track intrusion system and definition of a new set in correlation with outcomes of relevant S2R IP2 and IP5 projects

Final project progress report

Deliverables report on the main activities agreed in regular project meetings throughout the project lifetime

Report on evaluation of SMART2 prototype

This deliverable will report on all evaluation tests Task 63 and particularly on the results of the evaluation of integrated SMART2 OD and TDI prototype in realworld railway scenarios Task 64 The report will also provide update on requirements and specifications defined in WP 1 based on evaluation of developed technologies

Algorithms for vision-based obstacle and track intrusion detection

Deliverable reports on the outcomes of Task 22 the algorithms developed for ODTDI using onboard sensors and the results of implementation of these algorithms implemented on sensor data recorded in different SMART2 static and dynamic field tests

Datasets for obstacle detection in railways

This deliverable will report on data sets generation Task 61 which will be used to support development and offline evaluation of OD and TID methods as well as DSS and interfaces

Report on functional testing of fully integrated multi-modal obstacle and track intrusion detection system

Deliverable reports on the outcomes of Task 53 on testing of fully integrated SMART2 prototype to be allowed for its use in WP6 evaluation activities

Freight specific use cases for obstacle detection and track intrusion systems

Deliverable reports on the outcomes of Task 1.1 (the analysis and definition of relevant freight specific use cases), which would be considered in further SMART2 activities

Analysis of requirements and definition of specifications for obstacle detection and track intrusion systems

Deliverable reports on the outcomes of the Task 1.2 (RAMS-related requirements), Task 1.3 (technical requirements) and Task 1.4 (technical specifications for all potential sub-systems and their interfaces), which will be considered in further SMART2 activities.

Other (1)

SMART2 public website and communication plan

Deliverable reports on public website set-up, on the project leaflets describing project objectives and results and on project communication plan to increase the project impact

Publications

Conference proceedings (5)

"Determining the Risk Acceptance Criteria for Operational Changes In The Railway System Using The ""SAFETY II"" Principle"

Auteurs: Rosic S., Mitrovic M., Stamenkovic D.
Publié dans: Proceedings of the 19th International Scientific-Expert Conference on Railways - RAILCON 2020, 2020, Page(s) Page 117-120
Éditeur: University of Nis, Faculty of Mechanical Engineering

Deep learning-based rail track detection

Auteurs: Marten Franke, Danijela Ristić-Durrant, Kai Michels
Publié dans: XX International Scientific-expert Conference on Railways – RAILCON 2022, 2022, Page(s) Page 33-36
Éditeur: University of Nis, Faculty of Mechanical Engineering

Towards holistic autonomous obstacle detection in railways by complementing of on-board vision with UAV-based object localization

Auteurs: Franke, M., Reddy, C., Ristić-Durrant, D., Jayawardana, J., Michels, K., Banić, M., Simonović, M.
Publié dans: 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2022, ISBN 978-1-6654-7927-1
Éditeur: IEEE

Bounding Box Dataset Augmentation for Long-range Object Distance Estimation

Auteurs: Franke, M., Gopinath, V., Reddy, C., Ristić-Durrant, D., Michels, K.
Publié dans: International Conference on Computer Vision Workshops (ICCV Workshops), 2021, ISBN 978-1-6654-0191-3
Éditeur: IEEE

RAILWAY INFRASTRUCTURE MONITORING USING CLASSICAL COMPUTER VISION AND CONVOLUTIONAL NEURAL NETWORKS

Auteurs: Petrović, A.D.,Simonović, M., Banić, M., Stan S.-D.
Publié dans: XX International Scientific-expert Conference on Railways – RAILCON 2022, 2022, Page(s) Page 129-132
Éditeur: University of Nis, Faculty of Mechanical Engineering

Peer reviewed articles (7)

Deep learning-based obstacle detection and distance estimation using object bounding box

Auteurs: Ristić-Durrant, D., Franke, M., Michels, K., Nikolić, V., Banić, M., Simonović, M.
Publié dans: Facta Universitatis, Series: Automatic Control and Robotics, Numéro Vol 20, No. 2, 2021, Page(s) Page 75-85, ISSN 1820-6417
Éditeur: University of Nis

Object-Level Data Augmentation for Deep Learning-Based Obstacle Detection in Railways

Auteurs: Franke, M., Gopinath, V., Ristić-Durrant, D., Michels, K.
Publié dans: Applied Sciences, Numéro Appl. Sci. 2022, 12, 2022, ISSN 2076-3417
Éditeur: MDPI
DOI: 10.3390/app122010625

Use cases for obstacle detection and track intrusion detection systems in the context of new generation of railway traffic management systems

Auteurs: Hyde, P., Ulianov, C., Liu, J., Banić, M., Simonović, M., Ristić-Durrant, D.
Publié dans: Proceedings of the Institution of Mechanical Engineers, Part F: Journal of Rail and Rapid Transit, Numéro Volume 236, Numéro 2, February 2022, 2022, ISSN 0954-4097
Éditeur: Mechanical Engineering Publications Ltd.
DOI: 10.1177/09544097211041020

Dealing with Low Quality Images in Railway Obstacle Detection System

Auteurs: Perić, S., Milojković, M., Stan, S.-D., Banić, M., Antić, D.
Publié dans: Applied SCiences, Numéro Appl. Sci. 2022, 12, 3041, 2022, ISSN 2076-3417
Éditeur: MDPI
DOI: 10.3390/app12063041

Integration of Computer Vision and Convolutional Neural Networks in the System for Detection of Rail Track and Signals on the Railway

Auteurs: Petrović, A.D., Banić, M., Simonović, M., Stamenković, D., Miltenović, A., Adamović, G., Rangelov, D.
Publié dans: Applied Sciences, Numéro Appl. Sci. 2022, 12, 6045, 2022, ISSN 2076-3417
Éditeur: MDPI
DOI: 10.3390/app12126045

A Review of Vision-Based On-Board Obstacle Detection and Distance Estimation in Railways

Auteurs: Danijela Ristić-Durrant, Marten Franke, Kai Michels
Publié dans: Sensors, Numéro 21/10, 2021, Page(s) 3452, ISSN 1424-8220
Éditeur: Multidisciplinary Digital Publishing Institute (MDPI)
DOI: 10.3390/s21103452

Analysis of the Safety Level of Obstacle Detection in Autonomous Railway Vehicles

Auteurs: Rosić, S., Stamenković, D., Banić, M., Simonović, M., Ristić-Durrant, D., Ulianov, C.
Publié dans: Acta Polytechnica Hungarica, Numéro Vol. 19, No. 3, 2022, 2022, Page(s) Page 187-205, ISSN 1785-8860
Éditeur: Obuda University
DOI: 10.12700/aph.19.3.2022.3.15