Upgrade to Pro — share decks privately, control downloads, hide ads and more …

AI Agents for Earth Observation

AI Agents for Earth Observation

Shoaib Burq

June 21, 2026

Transcript

19 slides
  1. Agentic AI for Earth Observation S. Burq & M. Hassan

    | Decision Labs S. Burq & M. Hassan (decision-labs.com) | Big Data from Space 2025, Riga, Latvia
  2. How Agentic AI Works: A Primer for Earth Observation Goal:

    - Introduce the concept of agentic AI - Explain how agents work - Provide examples of EO-related agents - Call to action for collaboration on Open EO agents S. Burq & M. Hassan (decision-labs.com) | Big Data from Space 2025, Riga, Latvia
  3. Introduction to Agentic AI Hold on to your hats, because

    we're going to be introducing some new terms! ‣ Tools ‣ Agents ‣ Workflows ‣ Memory ‣ Evals S. Burq & M. Hassan (decision-labs.com) | Big Data from Space 2025, Riga, Latvia
  4. What is an agent? S. Burq & M. Hassan (decision-labs.com)

    | Big Data from Space 2025, Riga, Latvia
  5. "I am in Riga today can you suggest some places

    to visit and what should I wear?" S. Burq & M. Hassan (decision-labs.com) | Big Data from Space 2025, Riga, Latvia
  6. ! Agent's Toolbox: ‣ geocode-destination() → Get coordinates for the

    destination city ‣ get-weather-info() → Get weather information for the destination ‣ find-attractions() → Find attractions and points of interest in the destination ‣ create-itinerary() → Create daily itinerary with optimized routes S. Burq & M. Hassan (decision-labs.com) | Big Data from Space 2025, Riga, Latvia
  7. Travel Planner Workflow geocode- destination() get-weather- info() find- attractions() create-

    itinerary() generate-tour- summary() The LLM is used by the agent to plan the workflow and call the tools in the correct order. S. Burq & M. Hassan (decision-labs.com) | Big Data from Space 2025, Riga, Latvia
  8. ! EO Related Agent Examples ‣ Catalog Agent: Searches satellite

    archives (Copernicus, Sentinel, Landsat) ‣ Database Agent: Queries PostGIS / vector DBs ‣ File Agent: Loads GeoJSON, shapefiles, NetCDF datasets ‣ Imagery Agent: Processes SAR, Multi-spectral, Hyperspectral ‣ more ... S. Burq & M. Hassan (decision-labs.com) | Big Data from Space 2025, Riga, Latvia
  9. ! EO Related Agent Examples ‣ Forecasting Agent: Predicts weather,

    climate, crop yield using ML/ARIMA ‣ Visualization Agent: Plots maps (Deck.gl, Leaflet, Mapbox, Kepler.gl) ‣ Vision Agent: Detects objects in images (cars, buildings, etc) S. Burq & M. Hassan (decision-labs.com) | Big Data from Space 2025, Riga, Latvia
  10. Tools = "arms and sensors" of an agent Types of

    Tools: - APIs: Geocoders, Routing, etc - Models: Vision, Multimodal, etc - Commands: gdal, ogr2ogr, etc - Data Access: STAC Catalogues, Vector DBs S. Burq & M. Hassan (decision-labs.com) | Big Data from Space 2025, Riga, Latvia
  11. Multi-Agent Topology 1 1 citation: Principle of building AI agents

    by Sam Bhagwat S. Burq & M. Hassan (decision-labs.com) | Big Data from Space 2025, Riga, Latvia
  12. ! Maritime & Fisheries Workflow: Illegal Fishing Detection in the

    Mediterranean User asks: "Detect vessels fishing illegally in EU waters." Tools: ESA Sentinel-1 SAR → Vessel detection. AIS Data Integration → Match vessels with AIS signals. Vector Analysis Tool → Cross-check with legal fishing zones. Routing Tool → Closest coast guard patrols. Output: Alerts + patrol coordination for illegal fishing detection. S. Burq & M. Hassan (decision-labs.com) | Big Data from Space 2025, Riga, Latvia
  13. Workflow for Maritime & Fisheries ESA Sentinel-1 SAR AIS Data

    Integration Vector Analysis Tool Routing Tool S. Burq & M. Hassan (decision-labs.com) | Big Data from Space 2025, Riga, Latvia
  14. Conclusion: What have we learned? ‣ What is an agent?

    ‣ What are tools? ‣ What is a workflow? ‣ What is a multi-agent system? ‣ What are examples of Earth Observation Agents? S. Burq & M. Hassan (decision-labs.com) | Big Data from Space 2025, Riga, Latvia
  15. Thank you & For More Join the News Letter S.

    Burq & M. Hassan (decision-labs.com) | Big Data from Space 2025, Riga, Latvia
  16. ‣ Context Length ‣ MCP, A2A, RAG, Vector DBs ‣

    Multi-Agent Systems S. Burq & M. Hassan (decision-labs.com) | Big Data from Space 2025, Riga, Latvia

More Decks by Shoaib Burq

Other Decks in Science