AI for Space Competence Centre
Artificial intelligence is crucial for all space applications, but one of the most critical problems in this area is the enormous distances involved. The resulting time lag would in many cases render AI unusable with current technologies. This Centre of Competence is dedicated to developing innovative solutions to overcome this crucial challenge, making AI effective and usable in space despite significant latencies.


We imagine a future in which AI is able to operate autonomously in deep space, making critical decisions in real time without relying on communication with Earth. From the control of spacecraft on interplanetary missions to the management of lunar or Martian bases, we aim to create AI systems capable of operating effectively despite communication delays of minutes or even hours
Our research and development areas include:
- Distributed AI and Space Edge Computing: Development of AI architectures capable of operating locally on spacecraft and rovers, minimising dependence on communications with Earth.
Prediction and Anticipation Algorithms: Creation of AI models capable of predicting future scenarios and preparing responses in advance, compensating for communication delays.
Autonomous Decision Systems: Implementation of algorithms for autonomous decision-making under conditions of high uncertainty and limited communication.
AI Data Compression and Optimisation: Development of techniques to reduce the amount of data to be transmitted while maintaining AI effectiveness despite limited bandwidth
Interruption-Resilient AI: Creating AI systems capable of maintaining critical functionality even during prolonged interruptions in communication with Earth.
Earth-Space AI synchronisation: Development of methods to keep AI models on Earth and in space aligned, despite update latencies.
Advanced Simulation for Space AI: Creating test environments that accurately replicate the latency and uncertainty conditions of deep space for training and validation of AI systems.
