Flying Wings: National Defence Hackathon
Autonomous Tactical Quadcopter for GPS-Denied Subterranean & Recon Operations
Performance Metrics
Bench-validated telemetry benchmarks, endurance parameters, and operational bounds.
Ground speed in autonomous cruise mode
Continuous flight under full sensor load
Line-of-sight telemetry & autonomous patrol range
Including battery, avionics & perception payload
Autonomous navigation, obstacle avoidance & safe touchdown
GPS-denied visual-inertial odometry cumulative drift
Operational Blueprint
Developed as a high-stakes entry for the National Defence Hackathon, the Flying Wings platform is engineered specifically to operate in electronic warfare environments where satellite signals (GPS, GLONASS, NavIC) are jammed, spoofed, or physically degraded.
The airframe is built around a rigid carbon-fiber monocoque chassis integrating high-efficiency brushless propulsion, an edge AI perception payload driven by NVIDIA Jetson, and an onboard stereo-vision VIO (Visual-Inertial Odometry) pipeline that tracks millimeter-scale drift without external telemetry.
Real-time object detection models running on TensorRT identify tactical markers, structural breach points, and personnel zones, executing search-and-survey patterns autonomously while preserving complete radio silence.
Zero-GPS Odometry Fusion
Maintain positional hold and waypoint navigation using fused visual-inertial odometry and downward optical flow inside enclosed mock defense structures.
Target Recognition & Threat Tagging
Deploy deep neural networks on the Jetson Orin Nano to identify human subjects, hazards, and mission targets with sub-second inference latency.
Emergency Autonomous Touchdown
In the event of motor failure or payload disruption, compute optimal safe landing coordinates using real-time depth topology maps.
Low-Observability Radio Protocol
Execute tactical search sweeps while suppressing active telemetry broadcasts to evade electronic detection.
Avionics & Hardware Matrix
Subsystem flight controllers, edge companion compute, perception sensors, and telemetry infrastructure.
Flight Logs & Trials
Chronological deployment cadence, laboratory test phases, and verified field outcomes.
Agricultural Drone Collaboration (Georgia University)
