AI for Space
![Sentinel-2 false-color composite of Fyli, Athens in Greece. Before and after of the August wildfires. It contains modified Copernicus data. Processed by Sentinel Hub, acquired on 2023–08–23 and 2023–08–28.](https://www.ohb-hellas.gr/wp-content/uploads/2023/10/Sentinel-2-L2A-image-on-2023-08-23.jpg)
![Sentinel-2 false-color composite of Fyli, Athens in Greece. Before and after of the August wildfires. It contains modified Copernicus data. Processed by Sentinel Hub, acquired on 2023–08–23 and 2023–08–28.](https://www.ohb-hellas.gr/wp-content/uploads/2023/10/Sentinel-2-L2A-image-on-2023-08-28.jpg)
Efficiency: AI can automate numerous satellite operations, reducing the need for human intervention and potential human error.
-
Example: Have you ever gotten irritated over a slow internet connection? Novel AI Reinforcement Learning algorithms are being developed in order to improve just that! They work by predicting demand as the satellite flies over an area, taking into account factors like whether that area is a city or a rural environment, or the time of day and whether people are sleeping below. Based on those predictions they allocate the available spectrum (the frequencies in which a signal is transmitted down to earth) more optimally [1], resulting in much faster communication.
-
Real-time Decision Making: Satellites equipped with AI can make on-the-fly decisions.
-
-
Examples: They can adjust their imaging parameters based on cloud cover or other atmospheric conditions. They can use Cloud Detection models that have been tested in space-representative hardware, to adjust parameters and sensors, or even to save bandwidth and not send unusable images down to earth. Bandwidth is a precious commodity in spacecraft, because of the difficulty of sending data across such large distances quickly and with low error rates. Another example of autonomous decision making with ML models is the capability to classify images based on their usability based a classification model. Then we can use a ML compression model to reduce the data size, that varies based on what information we want to preserve. For example if we want to track forest fires we don’t care about very high resolution images in urban areas, and that saves precious time on the data transmission [2]. This can be extremely useful in situations that rely in real-time observations such as flood detection, fire detection or military applications.
-
Space Exploration: Rovers like NASA’s Perseverance use AI for tasks like autonomous driving and selecting scientific points of interest.
-
-
Example: NASA’s Perseverance Rover is dependent on AI self-driving models. The communication delay between Mars and Earth can be up to 22 minutes [3]. Given that delay, trying to drive a rover a distance of a few meters might take days. Imaging trying to drive your car, and for each input to your steering wheel (and seeing the road in front of you), you would have had to wait 22 minutes each way. Now imagine if the road was full of boulders and you would have to swerve every few meters to avoid them! So being AI-enabled [4], allows the rover to make real-time decisions, avoid obstacles, and cover more ground efficiently, maximizing the scientific return from the mission.
-
- Combining critical systems with capable systems. Often systems that offer high reliability (e.g. space-grade FPGAs with Real-Time Operating Systems) lack the computational capability that are available to us here on earth [1]. So the need has surfaced to combine a reliable system that handles critical operations with a capable system that meets the requirements.
- Harsh Environments: Space environments can be unpredictable. AI can adapt in real-time to situations like solar flares or unexpected satellite movements [2].
- Computational Limitations: Space hardware is optimized for durability, and not necessarily computational power. This poses challenges in running advanced AI models [3].
- Emerging Tech Hub: There is a growing tech and startup ecosystem in Greece, with a focus on cutting-edge technologies, including AI. But apart from startups, big tech companies also invest heavily in the country (Google, Amazon, Microsoft, etc.) with offices, data centers etc., solidifying the global confidence in Greece [1].
- Greece benefits from a scientifically-advanced as well as cost-effective talent pool that stands out in comparison to many other European nations.
- Given recent catastrophes, Greece has a need for near-real-time methods for fire detection, flood detection and maritime monitoring.
- Vision and Mission: We can offer a unique value proposition leveraging our relationship with Large European Space Integrators and advance our mission in making Greece a center of excellence in Space Software.
- Expertise: Specific expertise in Earth Observation (MFSR, AFD), and collaborations with relevant partners, positions OHB Hellas as an innovator.
- Success Stories: OHB Hellas is a Space Hub in Greece (LoIs, MoUs, contracts with local actors). It also is an extension of OHB SE‘s supply chain in Greece and remains strongly committed to continue development together with the local Greek Space ecosystem.
The universe is vast, mysterious, and holds countless secrets. While our ancestors looked up at the skies with wonder, today we have the tools, like AI, to take us a step closer to answering some of those age-old questions. Greece, steeped in tales of gods and stars, is not just reminiscing about its past but is actively shaping the future of space exploration. And at the heart of it all is OHB Hellas, not as mere spectators but as pioneers. So, as we stand at this exciting crossroad, we invite you to join us; because sometimes, to move forward, we need to look up.
Below you can find some projects we have accomplished in the field of AI/ML in space:
- Super Resolution
- Fire Detection
- Compression and Classification
- Satellite Communications Optimization
- Cloud Detection
- Vessel Detection
- Flood Detection
Furthermore, detailed examples of our work can be found in our Publications. We recommend our latest paper on one of the best ways to run AI/ML models on-board satellites!
Featured image: Sentinel-2 false-color composite of Fyli, Athens in Greece. Before and after of the August wildfires. It contains modified Copernicus data. Processed by Sentinel Hub, acquired on 2023–08–23 and 2023–08–28.