Deliverables

The purpose of the Quality and Risk Plan is to provide detailed process requirements with respect to the day-to-day project management and on the particular practical aspects of the project development, including report production procedures, criteria for work progression and results performance measurement. This deliverable shall be used by all the project partners in order to ensure quality assurance of project processes and outputs and prevent possible deviations from the project work plan.

A DMP describes the data management life cycle for the data to be collected, processed and/or generated by a project.

This deliverable provides a description of the meaning of the Ground Truth and the Digital Map concepts and of the user needs related to them. Such user needs have been contextualized in the context of the ERTMS train position and odometry principles and the interoperability performance requirements specified in Subset 041. Some practical use cases that can get benefits from the Ground Truth and the Digital Map have been also described with qualitative and quantitative examples.

The contents of this document includes the design of the High Level Logical Functional Architecture of the RAILGAP infrastructure capable of meeting the user requirements identified during WP2.1 activities and reported in RAILGAP D2.1.

Specification of the Measurement Information” provides the description of the measurement information to be acquired from each sensor and technology selected to be part of the RAILGAP Infrastructure and of the procedures to be adopted for executing the measurements. It also provides indications on the types of measurement equipment selected.

Specification and Description of the Measurement Infrastructure” provides a general overview of the RAILGAP measurement environment and a detailed description of the measurement configuration settings and measurement procedures to collect the measurement information introduced into the deliverable D3.1.

This deliverable provides the Electrical and Mechanical Design of the RAILGAP measurement infrastructure. Starting with the analysis of the main relevant standards applicable in the european railway context, the deliverable reports the electrical and mechanical drawings that have been realized to support the installation of the RAILGAP on-board measurement subsystem on the three different measurement trains (i.e. two in Italy and one in Spain)

The deliverable summarizes the performed field test campaigns outlining the travelled kms, the test periods, the configurations of the measurement infrastructure adopted during the field test campaigns, and the references to identify the stored measurements in the repository.

The document contains first introduction to each of the sensor technologies that are considered in RAILGAP and the key important elements for the use of the sensors for different use cases related to localization systems and for the main goals of RAILGAP on the elaboration of railway ground truth and digital map generations. This deliverable covers different aspects, error isolation techniques, modeling methodologies and initial results for some of the sensors in this first phase.

This deliverable discusses potential sensor Fault Detection and Discard (FDD) methods suitable for the Railgap architecture. It represents the first of three layers of fault protection. By applying these FDDs, the probability of sensor faults for GNSS, IMU, Lidar and camera is limited and the input for the sensor fusion algorithm is protected.

The deliverable provides an overview of suitable state of the art fault detection and exclusion (FDE) algorithms and strategies at the sensor fusion level. The FDE algorithms provide two functionalities: first, to detect the presence of a fault or an unexpectable error, and second, to identify and exclude all measurements containing fault(s) with a high confidence

Deliverable D4.4 summarizes the designs and relevant methods to perform the multi-sensor fusion. This deliverable is focused on the general fusion algorithms considering that sensor fault detection, discard algorithms and multi-sensor fault detection and exclusion techniques developed in other activities of the project are applied, providing a more robust and reliable sensor fusion solution.

The deliverable details the work carried out in the project related to the error modeling of different sensor technologies that are used for a robust and reliable positioning, trajectory determination and mapping in railway scenarios. It is an extension to deliverable D4.1 and includes updates considering a larger amount of data collected in the field

The deliverable provides an overview of possible fault detection and discard (FDD) algorithms and strategies based on state-of-the-art (SOTA) technologies

The deliverable provides an overview of suitable fault detection and exclusion (FDE) algorithms and strategies at the sensor fusion level. The FDE algorithms provide two functionalities: first, to detect the presence of a fault, and second, to identify and exclude all measurements containing fault(s) with a requested reliability

The deliverable summarizes the designs and relevant methods to perform the high-accuracy and high-reliability trajectory determination. The document not only describes the multi-sensor fusion and multi-layer protection architecture, but also detail on the proposed sensor fusion process and how to provide in the end the corresponding confidence intervals

This deliverable defines the architectural characteristics and the the main functionalities related to the functional blocks of the RAILGAP infrastructure: the On-Board Measurement (OBM) Unit and the Trackside Data Collector (TDC) Unit, also called Trackside Diagnostic Collector. Furthermore it defines the interfaces between OBM and TDC and between TDC and the data consumers (i.e., Technology Characterization, Digital Map, and Ground Truth).

This deliverable specifies the requirements of the On-Board Measurement (OBM) Unit and outlines the equipment implemented on the trial vehicles.

This deliverable specifies the requirements of the Trackside Data Collector (TDC) Unit, also called Trackside Diagnostic Collector. It also outlines Mission Log Files naming scheme and the file system structure used for their storage in the data repository.

This deliverable describes the development of the Measurement infrastructure that include (a) both the on-board measurement environment and the Trackside Central Collector, (b) the configurations required for allowing its installation and put in service and (c) the measurement procedures required for collecting measurements in accordance with the specification. This task also addresses the description and the setup of the facilities for enabling the transfers of the measurements from the on-board measurement environment to the Trackside Central Collector service and (c) the measurement procedures required for collecting measurements in accordance with the specification.

This document represents the Deliverable D5.5 “Description of the Diagnostic Recording Unit (concept Phase), This task is aimed at the definition (concept phase) of an on-board low-cost Diagnostic Recording Unit (DRU) that would meet the measurement environment requirements applicable for massive installations; this concept phase will thus allow for future developments in other projects. The activity of TP5.5 started from the requirements and the architecture defined in the documents listed in references to propose a more generic, useful and architecture for the DRU functionality/sub-system.

The document contains the details of the Ground Truth concept definition, in terms of variables, Key performance Indicators, and Minimum expected performances. A preliminary approach to the architectural design is also provided, with the associated requirements for the Data Fusion Algorithm, Ground Truth Construction, Output Management and User Interface.

The document includes the specifications of the Ground Truth toolset according to the logical and functional architecture and the performance requirements identified at Ground Truth concept level.

This deliverable includes an overview of the GT toolset and its functionality. It also contains general information of the work logic flown until the development of the toolset and the reference to the related documents prepared along the project. Results and performance achieved for a selected set of data as shown during the Malaga demo meeting held on the 19 June 2024 are also included.

This deliverable includes the records from the validation phase, including: • The description of the methodology applied and the acceptance criteria. • Internal Verification records • The description of the selected scenario • A summary of the achieved performances • The compliance matrix for the GT toolset requirements.

The present document contains the details of the Ground Truth concept definition, in terms of variables, Key performance Indicators, and Minimum expected performances. An approach to the architectural design is also provided, with the associated requirements for the Data Fusion Algorithm, Ground Truth Construction, Output Management and User Interface

This deliverables aims at 1. Defining the functional and performance requirements of trackside digital map in terms of accuracy, associated confidence interval and errors. 2. Specifying trackside elements and conditions to be considered for the digital map compliance with the railway standardised format 3. Specifying sensor data/information to be used as input for each identified algorithm for digital map design and reconstruction. 4. Defining technical performance required by multi-sensor platform in order to consider train dynamic effects of vehicles on tracks 5. Performing a survey of state of art of RTK/PPP DFMC algorithms and used sensor technologies definition, swot analysis and identification of candidate algorithms able to exploit the collected data for building the digital map; 6. Studying the state of the art of the different geolocation technologies based on IMU, Lidar or Digital cameras, identify which can be effectively combined with RTK / PPP and study the algorithms that allow combining both information to improve positioning. These algorithms include techniques of AI and video analysis to identify objects that can be used as a reference to improve positioning locally.

This document aims at detailing the following activities: • design of identified algorithms like RTK/PPP and data fusion with other external sensors and inclusion of the information suggested by the FDD Algorithms fully tailored for digital map definition taking into account WP4 and WP6 outputs; • high level design of a self-learning system that allows to generate maps dynamically and that improves the position system and location of the different map elements by combining the information collected by the different sensors or / and positioning systems; • generation of Track Topology part of the Digital Map database with precise geographical coordinates for all the track segments and their correlation to the relative longitude from one of the track segment boundaries; • identification of techniques for: (a) environment recognition and acquisition of key points through mathematical techniques, (b) detection of objects through neural networks and (c) generation of mathematical methods for calculating distances through image analysis and LiDAR information. • Testing and verification methodology definition.

This deliverable includes an overview of the system designed to create the required trackside digital map. It also contains general information of the work logic flown until the development of the toolset and the reference to the related documents prepared along the project.

This deliverable includes an overview of the analysis of the system developed to create the required trackside digital map. It also contains general information of the work logic flown until the development of the toolset and the reference to the related documents prepared along the project. As it will be reported in the following, the system has been evaluated by quantitatively estimating the performance achievable on the public RAILSEM19 dataset . The system has been then applied on data collected in operative conditions along the Pizarra-Los Prados railway during February 2024.

This document aims at performing the following activities: • Definition of the functional and performance requirements of trackside digital map in terms of accuracy, associated confidence interval and errors, taking into account the system requirements identified in WP2; • Specification of trackside elements and conditions to be considered for the digital map compliance with the railway standardized format (e.g. RailML); • Specification of the sensor data/information to be used as input for each identified algorithm for digital map design and reconstruction; • Definition of the technical performance required by multi-sensor platform in order to consider train dynamic effects of vehicles on tracks; • Survey of state of art of RTK/PPP Dual Frequency, Multi Constellation (DFMC) algorithms and used sensor technologies definition, swot analysis and identification of candidate algorithms able to exploit the collected data for building the digital map; • State of the art of the different geolocation technologies based on IMU, Lidar or Digital cameras; identify which can be effectively combined with RTK / PPP and study the algorithms that allow combining both information to improve positioning. These algorithms include techniques of Artificial Intelligence (AI) and video analysis to identify objects that can be used as a reference to improve positioning locally.

This deliverable addresses the ethics issues which may raise within RAILGAP project, related to the Humans requirement. This deliverable describes data protection and ethical issues considered within the RAILGAP project.

This deliverable provides the information on the procedures that will be implemented for data collection, storage, protection, retention and destruction of field measurement data during the field campaigns.