2nd Place International ERC Drone Task: Fully Autonomous UAV for Vision-Based Sample Localization & Precision Landing
2nd place internationally in the ERC drone-task scoring with 295/300 points, behind EPFL Xplore and tied with Brno Mars Rover, without GPS or GNSS: LiDAR, an optical-flow sensor and a Raspberry Pi 5 running the real-time CNN perception and MAVLink autonomous flight software I designed and implemented onboard.
- Three consecutive autonomous flights over the drone Mars Yard.
- 8/9 sample tubes detected and localized with ground-position accuracy below 10 cm.