Amazon's SageMaker AI is transforming how space agencies process the vast amounts of data generated during space missions. The company announced on June 26, 2025, that its Random Cut Forest (RCF) algorithm is being used by NASA and Blue Origin to detect anomalies in spacecraft dynamics data from lunar missions.
The collaboration specifically focuses on analyzing data from NASA and Blue Origin's demonstration of lunar Deorbit, Descent, and Landing Sensors (BODDL-TP). This unsupervised machine learning algorithm identifies unusual patterns in spacecraft position, velocity, and quaternion orientation data that might indicate critical moments during space operations.
"These anomalies most likely represent the lunar spacecraft vehicle dynamics at key maneuver stages of the deorbit, descent, and landing demonstration," according to Amazon's technical documentation. The technology can detect subtle deviations between data points while handling complex relationships between multiple parameters, making it particularly valuable for spacecraft monitoring.
The implementation uses Amazon's cloud infrastructure, with mission data stored in S3 buckets and processed through SageMaker AI's JupyterLab environment. Engineers train the RCF model using historical mission data, then deploy it to a scalable endpoint for ongoing anomaly detection.
This partnership comes at a significant time for Blue Origin, which is preparing to launch its Blue Moon Mark 1 lunar lander later this year. The insights gained from anomaly detection could prove crucial for ensuring mission success as both NASA and commercial space companies pursue increasingly ambitious lunar exploration goals.
By identifying anomalous data points that might otherwise be missed in the exponentially growing volume of telemetry data from space missions, Amazon's AI technology is helping to improve spacecraft health monitoring, engineering design, and mission planning for future space exploration.