AI-Enabled Air Traffic Control & Collision Prediction System

Personal Project | Python, AI | 2025

Overview

This project, named AIrspace, is a minimum viable prototype (MVP) of an AI-enabled air traffic control and collision warning system, designed as a response to the increasing number of near-miss incidents and the ongoing air traffic controller shortage in the United States.

Built in Python as a passion project, this early prototype simulates how a fully autonomous system could monitor aircraft in shared airspace, predict collision risks, and communicate advisories directly to pilots using AI-generated voice transmissions, mimicking a live air traffic controller.

Screenshot of AIrspace dashboard in normal state

Vision

The long-term goal is a system that can track every aircraft within a defined airspace and autonomously manage separation and traffic flow. The system would act as a real-time, intelligent assistant to, or full replacement for, human ATC operators in low-coverage or high-volume areas.

Key target capabilities include:

  • Real-time tracking of aircraft position, heading, altitude, velocity, and callsign

  • AI-generated voice communications transmitted over VHF to nearby aircraft

  • Predictive algorithms for mid-air collision avoidance

  • Scalable architecture for integration into high-traffic, multi-sector airspace

Current Prototype (MVP)

The MVP focuses on simulating a basic airspace with multiple aircraft and demonstrates core functionality, including:

  • Simulation of aircraft movement and speed & altitude data, with dynamic updates

  • Collision prediction algorithm based on vector intersection and altitude separation

  • Convergence warnings based on aircraft proximity and rerouting suggestions

Live telemetry, danger area (red), flight paths (blue), potential conflict warning system over NYC airspace

FLT2 leaving the danger areas of FLT3 & FLT7, proximity warning retracted

Future Development

Plans for continued development include:

  • Replace simulations with live aircraft tracking using ADS-B or FAA feed data

  • Implement AI voice outputs generated using text-to-speech to emulate real ATC communications

  • Explore feasibility of integration into existing air traffic control technologies

Motivation

This project was inspired by:

  • A growing number of near-miss incidents and runway incursions in 2023–2025

  • The critical shortage of air traffic controllers nationwide

  • The potential of AI systems to improve aviation safety

By prototyping this concept independently, I explored how automation and real-time AI decision-making could reduce human workload and improve reliability in safety-critical environments.