Periodic Reporting for period 3 - TAURO (Technologies for Autonomous Rail Operation)
Reporting period: 2023-01-01 to 2023-05-31
The EU-funded TAURO project of Shift2Rail contributed to shaping the future of European rail transport by researching technologies which will be required to make autonomous rail transport a reality. It achieved this by working on state-of-the-art systems for environmental perception, remote train operation, automatic monitoring and diagnostics, and facilitating the transition to automation from current train control systems.
The high-level objective of TAURO is to identify, analyse and finally propose suitable founding technologies for the future European automated and autonomous rail transport, to be further developed, certified and deployed through the activities planned for the Shift2Rail’s successor Europe’s Rail.
To achieve this, TAURO consortium broke down the work into several technical working areas, focusing on different system elements:
• Artificial environment perception for automation: AI is not deterministic and usual rail standards namely EN50126/8/9 cannot be used for certification as such. A rail specific approach is required.
• Remote driving and command: Ensure safety and security while providing performance and interoperable solutions.
• Automatic status monitoring and diagnostic for autonomous trains: Minimise human intervention to make autonomous trains autonomously resilient.
• Technologies supporting migration to ATO over ETCS: Accelerate the deployment of an interoperable ATO, enhance the maintainability and reliability of the delivered specifications and validate the stability and performance increase.
The project finalised in May 2023. Despite the challenging topics and short implementation time, the project achieved great results, producing relevant solutions and deliverables already transferred to Europe’s Rail.
The high-level objective of TAURO is to identify, analyse and finally propose suitable founding technologies for the future European automated and autonomous rail transport, to be further developed, certified and deployed through the activities planned for the Shift2Rail’s successor Europe’s Rail.
To achieve this, TAURO consortium broke down the work into several technical working areas, focusing on different system elements:
• Artificial environment perception for automation: AI is not deterministic and usual rail standards namely EN50126/8/9 cannot be used for certification as such. A rail specific approach is required.
• Remote driving and command: Ensure safety and security while providing performance and interoperable solutions.
• Automatic status monitoring and diagnostic for autonomous trains: Minimise human intervention to make autonomous trains autonomously resilient.
• Technologies supporting migration to ATO over ETCS: Accelerate the deployment of an interoperable ATO, enhance the maintainability and reliability of the delivered specifications and validate the stability and performance increase.
The project finalised in May 2023. Despite the challenging topics and short implementation time, the project achieved great results, producing relevant solutions and deliverables already transferred to Europe’s Rail.
The main goal for the final period (2022) was the completion of the activities and the dissemination of them, including the knowledge transfer to Europe’s Rail, and to its project FP2 R2DATO in particular.
Results were prolific and of hight value. They included (non-exhaustive list):
• Architecture specification for the remote driving, already being implemented in FP2 R2DATO
• The design and PoC of the common database for perception systems training, and a proposal for certification of such system
• This definition of a high number of use cases for perception systems in railway applications, including auto-diagnostics and security. For the latter a small demonstrator for interior security breach detection was implemented. Another relevant case was the feasibility studies on the applicability of the SLAM (Simultaneous Location and Mapping) for train location.
• Complementary work to X2Rail-4 on ATO GoA3/4 definition, including the lateral signal converter and the digital map modelling, and performing stability analysis of the ATO under ETCS for the TMS.
These results have been successfully integrated in FP2 R2DATO of Shift2Rail becoming an important starting baseline for the future developments, also in cooperation with the System Pillar.
Finally, during the final period, dissemination actions were reinforced, including the participation in WCRR, TRA, InnoTrans and the organisation of the final conference, jointly with X2Rail-4 and X2Rail-5.
Results were prolific and of hight value. They included (non-exhaustive list):
• Architecture specification for the remote driving, already being implemented in FP2 R2DATO
• The design and PoC of the common database for perception systems training, and a proposal for certification of such system
• This definition of a high number of use cases for perception systems in railway applications, including auto-diagnostics and security. For the latter a small demonstrator for interior security breach detection was implemented. Another relevant case was the feasibility studies on the applicability of the SLAM (Simultaneous Location and Mapping) for train location.
• Complementary work to X2Rail-4 on ATO GoA3/4 definition, including the lateral signal converter and the digital map modelling, and performing stability analysis of the ATO under ETCS for the TMS.
These results have been successfully integrated in FP2 R2DATO of Shift2Rail becoming an important starting baseline for the future developments, also in cooperation with the System Pillar.
Finally, during the final period, dissemination actions were reinforced, including the participation in WCRR, TRA, InnoTrans and the organisation of the final conference, jointly with X2Rail-4 and X2Rail-5.
Updated state-of-the-art:
During the execution of the European H2020 programme (2014-2022), many projects brought together stakeholders of the automotive industry for relevant safety and validation thematic areas. Many of them included particular developments or focus actions, but in general, they all contributed to the establishment of baseline know-how and de-facto standards in testing and validations methods for autonomous driving. To name a few, advances in data sharing and AI development were proposed and explored in ADAS&ME, AutoMate, AUTOPILOT, AVENUE, SHOW, BRAVE, CARAMEL, CoEXist, Dream4Cars, ENABLE-S3, ENSEMBLE, FABULOS, HEADSTART, ICT4CART, INFRAMIX, InLane, L3Pilot, MAVEn, Interact, SLAIN, TransAID, VI-DAS, 5G-Croco or 5G-MOBIX.
In parallel, at European level, other public and private initiatives have started to consolidate the research outcomes of these projects, including the EU expert group on Connected, Cooperative and Automated Mobility (CCAM), which, among other topics, aims to provide a Common Evaluation Methodology (CEM) and Data Sharing Framework (DSF) for Connected Automated Driving (CAD). Recently finished and ongoing projects are now pivoting around these concepts, such as the ARCADE Coordination and Support Action, which developed an online Knowledge Base (KB) gathering up to date information of CCAM related R&I projects, testing and piloting activities in Europe, regulation and policies, standards and impact assessment and data sharing methodologies. Currently, the FAME project aims to take over the results of ARCADE, and give continuity to the published KB. Focusing on scenario-based testing approaches, the Horizon Europe SUNRISE project aims to provide the first European-level Federated Scenario database for CAD testing.
Considering scientific initiatives, the most relevant contributions on AI development and testing are about the publication or massive multi-sensor datasets that can be used to train and validate models for environment perception (cameras, LIDAR, RADAR), and general scene understanding (e.g. maneuver analysis). In the last few years (2020-2022), it is noteworthy that large private companies are releasing professional-level datasets open for research purposes, such as the 3DHD CityScenes (Volkswagen), NuPlan (extension of NuScenes), MIT-AVT, Waymo Motion, Uber Pit30M, Ford Autonomous Vehicle Dataset, Drive&Act, DMD (Driver Monitoring Dataset), etc.
In the railway sector, two main datasets were recently released: FRSign, a dataset for traffic light detection and recognition, was published in 2020 as part of the TAS (Safe Autonomous Land Transport) project; Railsem19, a dataset for semantic rail scene segmentation, was published in 2019 on the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).
Since AI is the building block of most advances in CAD, there are also important trends on AI certification under the scope of Europe's AI strategy. Since its publication in 2018, several aspects of AI are being analyzed and recommendations are produced, such as the EU Open Data Directive, European Data Governance Act, EU AI Act, etc. Bodies such as the European AI Alliance have produced Ethics guidelines for trustworthy AI (2019), with operational tool assessment list for trustworthy AI (ALTAI, 2020), and general definition of best practices for AI practitioners.
During the execution of the European H2020 programme (2014-2022), many projects brought together stakeholders of the automotive industry for relevant safety and validation thematic areas. Many of them included particular developments or focus actions, but in general, they all contributed to the establishment of baseline know-how and de-facto standards in testing and validations methods for autonomous driving. To name a few, advances in data sharing and AI development were proposed and explored in ADAS&ME, AutoMate, AUTOPILOT, AVENUE, SHOW, BRAVE, CARAMEL, CoEXist, Dream4Cars, ENABLE-S3, ENSEMBLE, FABULOS, HEADSTART, ICT4CART, INFRAMIX, InLane, L3Pilot, MAVEn, Interact, SLAIN, TransAID, VI-DAS, 5G-Croco or 5G-MOBIX.
In parallel, at European level, other public and private initiatives have started to consolidate the research outcomes of these projects, including the EU expert group on Connected, Cooperative and Automated Mobility (CCAM), which, among other topics, aims to provide a Common Evaluation Methodology (CEM) and Data Sharing Framework (DSF) for Connected Automated Driving (CAD). Recently finished and ongoing projects are now pivoting around these concepts, such as the ARCADE Coordination and Support Action, which developed an online Knowledge Base (KB) gathering up to date information of CCAM related R&I projects, testing and piloting activities in Europe, regulation and policies, standards and impact assessment and data sharing methodologies. Currently, the FAME project aims to take over the results of ARCADE, and give continuity to the published KB. Focusing on scenario-based testing approaches, the Horizon Europe SUNRISE project aims to provide the first European-level Federated Scenario database for CAD testing.
Considering scientific initiatives, the most relevant contributions on AI development and testing are about the publication or massive multi-sensor datasets that can be used to train and validate models for environment perception (cameras, LIDAR, RADAR), and general scene understanding (e.g. maneuver analysis). In the last few years (2020-2022), it is noteworthy that large private companies are releasing professional-level datasets open for research purposes, such as the 3DHD CityScenes (Volkswagen), NuPlan (extension of NuScenes), MIT-AVT, Waymo Motion, Uber Pit30M, Ford Autonomous Vehicle Dataset, Drive&Act, DMD (Driver Monitoring Dataset), etc.
In the railway sector, two main datasets were recently released: FRSign, a dataset for traffic light detection and recognition, was published in 2020 as part of the TAS (Safe Autonomous Land Transport) project; Railsem19, a dataset for semantic rail scene segmentation, was published in 2019 on the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).
Since AI is the building block of most advances in CAD, there are also important trends on AI certification under the scope of Europe's AI strategy. Since its publication in 2018, several aspects of AI are being analyzed and recommendations are produced, such as the EU Open Data Directive, European Data Governance Act, EU AI Act, etc. Bodies such as the European AI Alliance have produced Ethics guidelines for trustworthy AI (2019), with operational tool assessment list for trustworthy AI (ALTAI, 2020), and general definition of best practices for AI practitioners.