Description du projet
Un outil intelligent pour les satellites
Une nouvelle génération de satellites entièrement flexibles est en train de conquérir le segment spatial du marché. Afin de contrôler les charges utiles flexibles des satellites et leurs homologues traditionnels configurables, le projet ATRIA, financé par l’UE, entend construire un outil générique et intelligent, appelé AI-PCS. Les techniques d’IA lui fourniront l’intelligence nécessaire pour décider de manière autonome de la configuration optimale des ressources satellitaires disponibles afin d’optimiser le service à la demande. Le processeur numérique transparent constituera le cœur de ces charges utiles numériques, leur offrant de nouvelles capacités dans le cadre des derniers paradigmes de réseau de la révolution 5G, qui n’auraient pas pu se concrétiser avec des charges utiles analogiques traditionnelles, telles que l’allocation adaptative de la bande passante et de la puissance, et la flexibilité de la connectivité et de la couverture.
Objectif
A new generation of completely flexible satellites in terms of mission definition is currently appearing in the space segment as a response to the 5G revolution, resulting in an unprecedented integration of the satellite services with the terrestrial deployments. In this context, the Digital Transparent Processor (DTP) is aimed to constitute the core of these digital payload, providing them with new capabilities unthinkable to achieve with the traditional analogue ones, such as adaptive bandwidth (BW) and power allocation, or flexibility in the connectivity and coverage.
As a consequence, the flexibility offered by these kind of satellites leads to the increased complexity in their management. Then, the optimum reconfiguration of these complex payloads to fulfil the users requirements is no longer an achievable task for a payload engineer due to the real-time constraints and the number of possibilities. To solve this, these satellites require new and intelligent tools capable of optimally allocating the satellite resources, while monitoring the interferences in such a dynamic scenario.
In this point, Artificial Intelligence/Machine Learning (AI/ML) algorithms appear to substitute the operator decisions and manual operations while configuring the payloads. In atria project, these algorithms will compose an AI/ML module in a Payload optimization Control System (AI-PCS). The algorithms are to be trained with both real data sets of former manual operations and synthetic data. Also, the algorithms will receive information from external services to provide them with inputs regarding meteorology, traffic, availability, etc. in order to achieve the optimum decisions.
AI-PCS is to be validated using both a flexible payload emulator and real satellite environments. AI-PCS is aimed to be a generic tool, transparent to the payload, providing the satcom operators with added value and turning it into a cost-effective solution for the future of the ground segment.
Champ scientifique
- natural sciencesearth and related environmental sciencesatmospheric sciencesmeteorology
- engineering and technologyelectrical engineering, electronic engineering, information engineeringinformation engineeringtelecommunicationstelecommunications networksmobile network5G
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringcontrol systems
- engineering and technologymechanical engineeringvehicle engineeringaerospace engineeringsatellite technology
- natural sciencescomputer and information sciencesartificial intelligencemachine learning
Mots‑clés
Programme(s)
Régime de financement
RIA - Research and Innovation actionCoordinateur
28760 Tres Cantos
Espagne