For the past few months I’ve been building a ground control station app for UAVs that has real-time AI object detection and tracking built directly into the flight interface — no more tabbing between GCS and a separate vision tool, no more manually spotting targets in a video feed.
What it does:
Real-time object detection & tracking overlaid directly on the live video feed
Edge-AI controller with adjustable confidence threshold
Target lock/track with live detection log (person, vehicle, etc.)
Standard GCS features: ARM, flight planning, telemetry, camera link config
Swarm control support + automated 3D flight planning
Built this because every time I tried combining an off-the-shelf GCS with a separate detection pipeline, latency and integration issues killed the workflow. Wanted something where detection and control live in the same loop.
Demo video below shows it running live — detecting and tracking a person in real time while flying.
Screenshots of the interface attached too.
I ended up packaging the source code since a few people in similar projects asked if I’d share it — it’s available here if you want to use it as a base for your own GCS/AI integration: https://payhip.com/b/9Hx8p
Happy to answer any technical questions — framework, hardware compatibility, licensing, etc.