The mission of this UAV is to secure schools, businesses, and people using Computer Vision technology to determine human behaviour and posture. Using this information, it can track, follow, and understand people and what situation they may be in whether it be violent or just a normal interaction.
Specifics right now: Autonomous flight with following of a user for 12 minutes, a 2-3 foot by 6 foot flight distance from the user, and a 60% accuracy in determining the user's posture and safety.
Specs: Raspberry Pi 4, Python's OpenCV and Firebase for communications with Phone, Linux, CC3D Flight controller, Librepilot Firmware and a 3500 mAh LiPo battery.
Code is not yet opensource! (Needs improvements and retraining for reliability)
I participated in the CT-Stem Science Fair and the University of Bridgeport Fair!
Specs: BeagleBone Blue Ardupilot Board, Raspberry Pi 4, Python's OpenCV and FrSky Transmitter and Receiver for Communication)
This iteration aims to improve the EITS Drone's accuracy determining situations, reliability in autonomous flight, design + chasis, and battery life!