Descripción del proyecto
Mejorar la gestión del tráfico aéreo mediante la colaboración por aprendizaje automático en conjuntos de datos privados
La gestión del tráfico aéreo (GTA) puede beneficiarse enormemente de la explotación cibersegura de grandes conjuntos de datos privados que pertenecen a diferentes partes interesadas. Actualmente, sin embargo, existen reticencias en cuanto al hecho de compartir datos sensibles. El proyecto AICHAIN, financiado con fondos europeos, propone un innovador concepto de gestión de la información digital (GID) que ayudará a aprovechar estos valiosos conjuntos de datos privados. Combinará dos tecnologías emergentes de GID, el aprendizaje automático federado y la cadena de bloques, para definir un diseño de aprendizaje federado y avanzado que protege la privacidad y en el cual el intercambio de datos y de modelos de formación no se verá comprometido. También investigará los beneficios potenciales del innovador GID mediante casos de investigación sobre GTA relativos a la capacidad de demanda avanzada que equilibra un modelo pronóstico del gestor de redes reforzado con datos operativos reales de usuarios del espacio aéreo.
Objetivo
This project proposes an innovative Digital Information Management (DIM) concept, i.e. the AICHAIN solution, that aims at enabling the cyber-secured exploitation of large private data sets that belong to different stakeholders and that contain valuable information for ATM operations. To overcome the stakeholders’ reluctance to share sensitive data, the exploitation will not be performed by exchanging the data itself but by articulating an advanced privacy-preserving federated learning architecture in which neither the training data nor the training model need to be exposed. This will be possible thanks to the innovative combination of two emerging DIM technologies: Federated Machine Learning (FedML) and Blockchain technologies.
The potential benefits of the new proposed DIM concept will be explored through ATM research use cases related to advanced Demand Capacity Balancing (DCB) predictive models of the Network Manager (NM), whose prediction performance is expected to significantly improve thanks to the exploitation of relevant operational private data from Airspace Users. The accuracy of the new DCB predictive models augmented with real operational data accessed through the AICHAIN solution will be benchmarked against the machine learning models for DCB that are currently in use or under research by NM.
The project will also address the exploration of governance and incentives mechanisms as part of the AICHAIN solution concept architecture, to facilitate the adoption of the concept and the alignment of interests of the key stakeholders (especially of the data owners). The design of advanced governance & incentives mechanisms, which could be implemented using the mechanism of “smart contracts” available in the toolset of Blockchain, will be complemented with a theoretical identification of data exploitation benefits and with discussions in workshops participated by external experts.
Ámbito científico
Palabras clave
Programa(s)
Régimen de financiación
RIA - Research and Innovation actionCoordinador
08005 Barcelona
España