Earth observation satellite autonomously identifies targets using AI technology

Jun 15, 2026

AI Summary

An Earth observation satellite, Yam-9, successfully identified areas of interest on its own using a vision-language model developed by NASA. This milestone demonstrates the potential for AI to enhance satellite capabilities and reduce the data burden on ground analysts.

The Yam-9 satellite, built by Loft Orbital, autonomously found targets in April using AI technology without human intervention.
The software, developed by NASA's Jet Propulsion Laboratory, utilized Google DeepMind's Gemma 3 vision-language model to respond to natural language queries and classify sensor data.
This advancement could streamline data processing in space, allowing satellites to perform initial data triage and potentially enabling real-time monitoring capabilities.
Loft Orbital operates a business model focused on providing satellite infrastructure for third-party customers, with recent contracts including the launch of six satellites for EarthDaily.
The successful deployment of the vision-language model in orbit is expected to encourage other companies to explore similar AI applications in space.
Lessons learned from this deployment will inform future efforts to establish larger-scale AI infrastructure in space, particularly regarding power and memory management.
The concept of interactive AI assistants for astronauts is also being explored, inspired by the need for hands-free operation in space environments.

satelliteautonomousearth observationroboticsai technology

Earth observation satellite autonomously identifies targets using AI technology

Related Stories

Neolix's Executive Discusses Growth Strategy and Market Expansion

Robotics Progress Remains Slow Despite Industry Showcases

MIT researchers develop robotic boats for dynamic water infrastructure

Startup aims to revolutionize robotics with foundation models for AI