AeroIntel

AeroIntel is a real-time air traffic intelligence platform that automatically detects flight conflicts, analyzes congestion hotspots, and generates cost-optimized resolution proposals.

Comment

Inspiration

Air traffic management operates at an enormous scale, with planners responsible for coordinating thousands of flights across shared airspace every day. While safety is the primary concern, planners must also consider efficiency, delays, fuel costs, and downstream operational impacts.

While exploring modern air traffic and flight tracking systems, I noticed that many tools focus on displaying data, but few actively support decision-making. Conflict detection is often manual or rule-heavy, congestion patterns are difficult to interpret at scale, and resolution decisions rarely consider cost trade-offs.

I built AeroIntel with the goal of creating a decision-support co-pilot—a system that doesn’t just show where flights are, but actively analyzes trajectories, highlights risk early, and suggests optimized resolutions before problems occur.

What it does

AeroIntel is a real-time air traffic intelligence platform that transforms raw flight plans into actionable insights.

It allows me to:

  • Automatically detect flight conflicts by analyzing trajectories and identifying loss-of-separation events
  • Perform hotspot analysis to pinpoint high-density airspace regions that may lead to cascading conflicts
  • Generate intelligent resolution proposals, offering multiple options per conflict (altitude, speed, departure changes) ranked by cost-effectiveness
  • Visualize flights, conflicts, and hotspots on an interactive, rotatable 3D globe
  • Explore all insights through a Control Center interface that expands from a compact panel into a full-screen operational dashboard

How I built it

I designed and implemented AeroIntel entirely as a solo project, covering both the backend analysis and the frontend experience.

I built:

  • A synthetic flight plan ingestion pipeline using structured JSON
  • Trajectory interpolation to model aircraft positions over time
  • Conflict detection logic based on horizontal and vertical separation thresholds
  • Airspace cell aggregation for hotspot detection and scoring
  • Resolution generation and cost-aware ranking for conflict mitigation
  • An interactive 3D globe visualization
  • A responsive Control Center UI with expandable layouts and smooth transitions

I focused heavily on performance, modularity, and clarity to ensure the system remains responsive even when analyzing large numbers of flights.

Challenges I ran into

  • Balancing realism and performance while simulating and analyzing thousands of trajectories
  • Avoiding visual overload when presenting dense airspace data in a way that supports decision-making
  • Designing cost-aware decision logic that reflects operational trade-offs without excessive complexity
  • Managing end-to-end development as a solo builder, from system design to user experience

Accomplishments that I'm proud of

  • Building a complete end-to-end airspace intelligence platform entirely on my own
  • Analyzing 1000+ flights in seconds with automated conflict and hotspot detection
  • Designing a cost-aware resolution engine that presents multiple ranked options per conflict
  • Creating a FlightRadar24-inspired 3D globe interface with an expandable control center
  • Delivering a polished, interactive system focused on decision support rather than raw data display

What I learned

  • Visualization is most effective when it supports decision-making, not just exploration
  • Providing multiple ranked options leads to better human judgment than forcing a single resolution
  • Small UX details—motion, hierarchy, and transitions—significantly improve perceived system intelligence
  • AI systems are most powerful when they augment human expertise rather than replace it
  • Building solo strengthened my understanding across the full stack, from algorithms to user experience

What's next for AeroIntel

Next, I plan to:

  • Integrate real-world flight data sources
  • Incorporate weather and airspace restriction modeling
  • Improve proposal scoring using historical performance data
  • Add what-if simulations for delays and disruptions
  • Explore collaboration features for multi-user control centers

AeroIntel is designed to be extensible, and this project establishes a strong foundation for future research and real-world airspace planning applications.

Built With

Share this project:

Updates