Airport Layouts as Graphs: How I Built a Real-World Accurate, AI Air Traffic Control Agent
summary
Built an AI ATC agent that combines live METAR calls (via aviation.weather.gov), an open‑source runway selection algorithm from GitHub, and a Neo4j airport graph to compute exact taxi routes; deployed tools as webhooks and wrapped in an agent using Retool and CLA‑4. It reads METARs in pilot format, chooses the active runway (example result: runway 18), and returns precise taxi instructions (example: taxi to runway 18 via Charlie Alpha Alpha1), validated on Southeast Iowa Regional and an O’Hare scenario.
AI-written summary · about
chapters (8)
- 0:00Introduction and motivation
- 1:24Custom GPT limitations
- 2:49Key requirements for realism
- 5:35Accurate weather simulation
- 7:41Runway selection logic
- 8:46Airport as a graph database
- 10:49Agent architecture and demo
- 14:39Real-world application and summary
related talks
01h 30mHow to build AI Agents: A step by step guide to creating AI agents
40mAI Agents in Practice
53mHands-on with AI agents: From scratch to LangGraph
31mHow to Build Autonomous AI Agents That Run Anywhere
26mGrounded Reasoning Systems for Cloud Architecture
15mAI Studio - Creating and scaling your custom copilots
10mLightning Talk: AI4LREANET - AI for Critical Infrastructures
22mHOW TO BUILD YOUR OWN AI AGENT: THE COMPLETE ARCHITECTURE FOR ENGINEERING TEAMS