Airbus Defence and Space Uses Deep Learning and Google Cloud to Deliver Real-Time Satellite Imagery
Airbus Defence and Space· FranceGoogle Kubernetes Engine · Cloud Pub/Sub · Cloud Datastore +2Airbus Defence and Space's Intelligence business line rebuilt its satellite imagery platform, OneAtlas, on Google Cloud, using Kubernetes Engine, Cloud Pub/Sub, Cloud Storage and Cloud Datastore to serve more than 1,200 customers in over 100 countries. Airbus trained TensorFlow deep learning models to recognize cloud cover, cutting its cloud-detection error rate from 11% to 3%, and the new platform reduced the time for customers to access satellite imagery from hours to less than half a second while adding millions of square kilometers of imagery weekly.
Overview
Airbus Defence and Space's Intelligence business line rebuilt its satellite imagery platform, OneAtlas, on Google Cloud, using Kubernetes Engine, Cloud Pub/Sub, Cloud Storage and Cloud Datastore to serve more than 1,200 customers in over 100 countries. Airbus trained TensorFlow deep learning models to recognize cloud cover, cutting its cloud-detection error rate from 11% to 3%, and the new platform reduced the time for customers to access satellite imagery from hours to less than half a second while adding millions of square kilometers of imagery weekly.
This entry has 13 published fields tied to exact passages in an immutable source capture.
Inspect the highlighted sourceThe challenge
Airbus Intelligence's existing satellite imagery solution could not speed up production and delivery of image and data products fast enough to meet growing customer expectations; customers had to choose images from a catalogue and wait several hours to receive them, and the specialized imagery required customers to have significant technical capability to use it.
The solution
Airbus built a new platform called OneAtlas on Google Cloud, using Google Kubernetes Engine to orchestrate image-processing workloads and scale compute up and down with demand, Cloud Pub/Sub for messaging, Cloud Datastore and Cloud Storage to hold and serve imagery and data, and trained deep learning algorithms in TensorFlow to automatically recognize cloud cover in satellite imagery.
Sources
Open any source and check the claim yourself — that is the point of the register.
This record was researched and written with AI assistance, and its claims were checked against the sources above. (EU AI Act art. 50 transparency notice.)
Other aerospace & defense entries in the register.
Boosting Innovation and Cutting Costs Through Lockheed Martin's AI Factory
Lockheed Martin centralized compute resources, MLOps tools and best practices into the Lockheed Martin AI Factory, built on an NVIDIA DGX SuperPOD reference architecture, to build and deploy trustworthy AI at scale on-premises under strict data governance requirements. Developers can now get GPU-backed environments running in minutes instead of weeks, and training times dropped from weeks to days. The AI factory processes over one billion tokens per week and now serves 7,000 engineers and developers, supporting internal chatbots and coding assistants such as Lockheed Martin Text Navigator, and has consolidated 30+ models on-premises.
Lockheed MartinNVIDIA DGX SuperPOD · NVIDIA Triton Inference Server · NVIDIA NeMo Framework +2
Rolls-Royce uses AI turbine blade inspection and engine health monitoring to prevent 400 maintenance events a year
Rolls-Royce uses Microsoft Cloud for Manufacturing, Azure Databricks, and generative AI across engine design, turbine production, and engine health monitoring. Its 'Signature Analyzer' AI/machine learning system inspects turbine blade cooling holes (previously about 2 million inspected manually per month), using machine vibration analysis and generative AI to optimize defect detection, boosting machine utilization 30% and accelerating fault resolution from days to near real time. AI-powered engine health monitoring tracks more than 10,000 engine parameters and detects and prevents around 400 unplanned maintenance events a year, saving millions in repair costs.
Was this helpful?
Your feedback helps us improve our use case database