Redis 8 beyond caching to uniquely powerful traffic management system.

[MyrinNew]

This is a submission for the Redis AI Challenge: Beyond the Cache.


What I Built

A real-time air traffic control simulator that can track aircraft positions (latitude, longitude, altitude). Detect potential collisions using geospatial calculations, and visualize aircraft movement on a live map. It also generates alerts when aircraft violate safety thresholds.


Demo

Here is the GitHub repo: https://github.com/SebastianKibiriti...-with-deepseek


How I Used Redis 8

The features of Redis 8 that were used are: GEOADD, for storing the positions of the aircraft + GEORADIUS, for querying other aircrafts nearby.






# Store aircraft positions
redis.geoadd(
"aircraftositions",
{aircraft_id: (longitude, latitude)}
)

# Query nearby aircraft (1km radius)
conflicts = redis.georadius(
"aircraftositions",
current_lng, current_lat, 1, "km"
)







Lua Scripting for collision detection.






-- conflict_check.lua
local nearby = redis.call('GEORADIUS', KEYS[1], ARGV[1], ARGV[2], 1, 'km')
-- Checks altitude differences using ZSCORE







Streams, for real-time alerts. little to no alerts will be missed with faster replication.






# Trigger alert
redis.xadd(
"conflict_alerts",
{"trigger_id": "flight123", "conflicts": "flight456"}
)

# Frontend consumes via WebSocket
alerts = redis.xread({"conflict_alerts": "$"}, block=0)







Multithreaded I/O to optimize performance. This allows for more that 1000+ concurrent aircraft updates to be handled.


The combination of geospatial, streaming, and Lua scripting creates a unique traffic management system.


Architecture Diagram






More...