Periodic Reporting for period 1 - EMERALDS (Extreme-scale Urban Mobility Data Analytics as a Service)
Periodo di rendicontazione: 2023-01-01 al 2024-06-30
EMERALDS’s vision is to design, develop and create an urban data-oriented Mobility Analytics as a Service (MAaaS) toolset, consisting of the so-called ‘emeralds’ services, compiled in a proof-of-concept, capable of exploiting the untapped potential of extreme urban mobility data. The toolset will enable the stakeholders of the urban mobility ecosystem to collect and manage ubiquitous spatio-temporal data of high-volume, velocity and variety, analyse them both in online and offline settings, import them to real-time responsive AI/ML algorithms and visualize results in interactive dashboards, whilst implementing privacy preservation techniques at all data modalities and at all levels of its architecture. The toolset will offer advanced capabilities in data mining of large amounts and varieties of urban mobility data and its efficiency will be assessed, validated and demonstrated in three TRL 4 pilot use cases, and deployed/showcased in two early adopters’ data-driven TRL 5-6 applications.
EMERALDS’s vision will be realised through five (5) objectives: i) Design a service-oriented reference architecture of a palette of services (‘emeralds’) for extreme scale urban mobility data analytics, underpinned by a distributed computing environment ensuring that both edge and cloud processing contribute towards establishing a robust processing pipeline, ii) Develop extreme-scale acquisition and processing methods and tools for urban mobility data, which will be scalable with the data at hand, and at the same time facilitate accurate and low-latency data collection, pre-processing, mining, fusion, and management., iii) Develop mobility data analytics and AI/ML tools and services, appropriately designed to perform along the edge/fog/cloud continuum. iv) Demonstrate and measure the efficiencies of the services through three pilot use cases and validate the concepts and tools' usefulness through two early adoption applications, v) Develop and implement a plan for wider adoption and accelerate diffusion of innovation and exploitation of project outcomes through appropriate dissemination, communication, and exploitation activities.
EMERALDS’s vision will be realised through five (5) objectives: i) Design a service-oriented reference architecture of a palette of services (‘emeralds’) for extreme scale urban mobility data analytics, underpinned by a distributed computing environment ensuring that both edge and cloud processing contribute towards establishing a robust processing pipeline, ii) Develop extreme-scale acquisition and processing methods and tools for urban mobility data, which will be scalable with the data at hand, and at the same time facilitate accurate and low-latency data collection, pre-processing, mining, fusion, and management., iii) Develop mobility data analytics and AI/ML tools and services, appropriately designed to perform along the edge/fog/cloud continuum. iv) Demonstrate and measure the efficiencies of the services through three pilot use cases and validate the concepts and tools' usefulness through two early adoption applications, v) Develop and implement a plan for wider adoption and accelerate diffusion of innovation and exploitation of project outcomes through appropriate dissemination, communication, and exploitation activities.
During the first reporting period (M1-M18), EMERALDS has achieved the following:
a)Establishment of a Reference Architecture, based on literature, industry best practices, expert group and organization white papers (e.g. FIWARE, GAIA-X, BDVA and AIOTI), and different facets of mobility analytics
b)Creation of highly scalable, interoperable and reusable collection of ~30 software tools, offered as a Service, constituting the EMERALDS toolset capable of treating streaming and at rest data, live since July 2024 (T2.1 D2.2)
c)Extension and practical application of the computing continuum approach, enabling flexible and efficient application deployment by matching computational needs with available resources and investigation of the potential of edge computing for decentralized data processing
d)Enhanced interoperability through the development and promotion of the MovingFeatures and GeoParquet Standards
e)Developed scalable feature importance techniques for high-dimensional mobility data along with a vertical federated learning framework
f)Development of an MLOps platform tailored for extreme scale AI services, interoperable with the EU AI on Demand platform, providing a comprehensive environment for developers to manage the entire machine learning lifecycle, from data ingestion to model deployment
g)Ethical by design AI/ML algorithms under T4.2 D4.1 adhering to principles and relevant framework presented in D1.2 D7.1.
h)A set of cutting-edge security tools have been developed ensuring that nodes, communication channels, networks and data transfers running EMERALDS software are fully protected (T2.3).
i)Containerization of EMERALDS components paired to a robust CI/CD pipeline (D2.2 T2.1)
j)17 publications on the scientific and technical work carried out in scientific conferences and journals and participation in >20 events disseminating project results to the academic and broader community
k)Customisable analytics dashboard integrating WP4 services to the DigiTwin environment used for crowd management
l)Data cleaning, enrichment, processing and analytics pipelines designed for all EMERALDS use cases, combining services and WP2 horizontal tools
m)Complied a project-wide hierarchical KPI framework (KPI nexus), measuring software, business, use case technical and impact performance metrics utilized across WPs
a)Establishment of a Reference Architecture, based on literature, industry best practices, expert group and organization white papers (e.g. FIWARE, GAIA-X, BDVA and AIOTI), and different facets of mobility analytics
b)Creation of highly scalable, interoperable and reusable collection of ~30 software tools, offered as a Service, constituting the EMERALDS toolset capable of treating streaming and at rest data, live since July 2024 (T2.1 D2.2)
c)Extension and practical application of the computing continuum approach, enabling flexible and efficient application deployment by matching computational needs with available resources and investigation of the potential of edge computing for decentralized data processing
d)Enhanced interoperability through the development and promotion of the MovingFeatures and GeoParquet Standards
e)Developed scalable feature importance techniques for high-dimensional mobility data along with a vertical federated learning framework
f)Development of an MLOps platform tailored for extreme scale AI services, interoperable with the EU AI on Demand platform, providing a comprehensive environment for developers to manage the entire machine learning lifecycle, from data ingestion to model deployment
g)Ethical by design AI/ML algorithms under T4.2 D4.1 adhering to principles and relevant framework presented in D1.2 D7.1.
h)A set of cutting-edge security tools have been developed ensuring that nodes, communication channels, networks and data transfers running EMERALDS software are fully protected (T2.3).
i)Containerization of EMERALDS components paired to a robust CI/CD pipeline (D2.2 T2.1)
j)17 publications on the scientific and technical work carried out in scientific conferences and journals and participation in >20 events disseminating project results to the academic and broader community
k)Customisable analytics dashboard integrating WP4 services to the DigiTwin environment used for crowd management
l)Data cleaning, enrichment, processing and analytics pipelines designed for all EMERALDS use cases, combining services and WP2 horizontal tools
m)Complied a project-wide hierarchical KPI framework (KPI nexus), measuring software, business, use case technical and impact performance metrics utilized across WPs
EMERALDS brings together a consortium of leading research institutions, universities, innovative SMEs and industry partners to address the challenges of managing and analyzing large-scale mobility data. EMERALDS results bear the potential to create significant scientific, technological, social and economic benefits, pertaining to: a) stimulation of economic growth by creating new jobs and fostering innovation in the mobility sector b) enhancing interoperability, facilitating replicability, transferability and fast uptake of the EMERALDS Toolset c) improved operational efficiency such as reduced congestion, shorter travel times, and increased responsiveness in crowd management, multi-modal traffic management, and public transportation ridership; d) yielding cost savings , including reduced operational costs and increased revenue for transport network operators, city authorities and SMEs activated in the data economy; e) reduced emissions and improved air quality in urban environments due to efficient allocation and operation of mobility resources; f) privacy-preserving edge to cloud solutions respecting passengers anonymity applying the GDPR data minimisation principles g) introduce a dedicated field on mobility data science, including the development of new algorithms, techniques, and standards; h) enacting a pioneering role in driving technological innovation, such as the development of new hardware or software solutions and contributing to knowledge transfer and capacity building in the field of urban mobility data analytics; i) stimulating further research initiatives on Mobility Data Science expanding and adapting the EMERALDS solutions from constrained urban road networks, to peri-urban, airborne, waterborne and freight mobility; j) directions for improved urban planning and land use decisions that prioritize the right to mobility for all societal groups.