Skip to content
Swarm Drone Survivor Detection

Swarm Drone Survivor Detection

Eyes in the Sky, Hope on the ground.

Created on 18th May 2025

Swarm Drone Survivor Detection

Swarm Drone Survivor Detection

Eyes in the Sky, Hope on the ground.

The problem Swarm Drone Survivor Detection solves

When disaster strikes, finding survivors becomes a race against time. Large-scale earthquakes, floods, and building collapses create environments where traditional search methods fail. Rescuers face obstructed visibility from smoke and debris, collapsed infrastructure with no GPS or cellular networks, and vast unstructured terrain that's impossible to search thoroughly with current resources. Every minute counts, yet our ability to locate survivors in these critical moments remains dangerously limited.

Our solution to it:-
Smart drones detect disaster survivors using visual input and RF signals (e.g., Bluetooth, SOS beacons).
They operate as a coordinated swarm, adapting roles and search patterns in real-time.
The system ensures efficient coverage of complex, unstructured terrains.
Live maps visualize probable survivor locations and track drone movements.

Challenges we ran into

1:- link 1 down for SITL and MAVLINK integration, i struggled for 2 hours in that.
2:- Integration of various models of RF, YOLOv8
3:- Shortage of GPU space as we trained all models, here only in this hackathon.
4:- Integration of backeed files with frontend as we have to run tsx files separately.
5:- Establishing a communication between drones in the swarm.

Tracks Applied (1)

Ethereum Track

Project Overview: The AI-powered drone swarm simulates autonomous survivor detection in disaster scenarios (e.g., earthq...Read More
ETHIndia

ETHIndia

Discussion

Builders also viewed

See more projects on Devfolio