As drone technology grows rapidly, so too do accounts of drone misuse. The public have witnessed incidents in which drones have caused harm to the public sector, as well as to private entities. As such, we welcomed with enthusiasm an opportunity to contribute to the development of an autonomous drone hunter. In cooperation with Fly4Future, we have developed an autonomous anti-drone solution: Eagle.One.
Thanks to its cutting-edge control and computer vision techniques, the Eagle.one system is capable of fully autonomous drone hunting. The system has a wide range of applications, such as private protection from paparazzi drones and defense mechanisms for critical infrastructure, airports, government institutions, power stations, or stadiums. This fully autonomous robotic system is capable of 24/7 protection of individuals and their families, as well as protection of borders, jails, and prisons from deliveries of contraband by drones, and guarding public events against drone attacks.
Eagle.One drone hunter can easily be integrated into a comprehensive anti-drone solution. One solution tested used the Dedrone autonomous aerial intercepting system (AAIS) system, which detects helicopters in a no-fly zone using cameras and radio transmitters pre-installed throughout the area. Based on detected radio signals and image processing from the cameras, the system distinguishes a helicopter from birds or other objects. If a hostile helicopter flies into the no-fly zone, the Dedrone system detects it and sends a command to deploy an Eagle.One equipped with a net gun to capture the enemy drone.
In response to alerts, the drone hunter autonomously starts and flies towards the area of the estimated position of the hostile helicopter, while continuously receiving its updated position from the Dedrone system. Our defensive drone then uses its onboard cameras and other sensors to precisely track down the hostile helicopter. Onboard artificial intelligence is used for drone detection, classification, tracking, target trajectory estimation, movement prediction, and planning offensive maneuvers. With the target in focus using onboard artificial neural networks, the defensive drone shoots a net to capture the hostile helicopter. The offensive manoeuvre is triggered by an onboard computer gathering information from onboard sensors, although it can be preceded by a confirmation from the system’s operator. In order to speed up the maneuver, security critical applications can use the system solely in autonomous mode without any operator. After catching the helicopter, the net remains connected to the defensive drone via a rope, preventing the captured object from falling. It can then safely remove the target and land it at a designated area.
Publications:
Matouš Vrba, Yurii Stasinchuk, Tomáš Báča, Vojtěch Spurný, Matěj Petrlík, Daniel Heřt, David Žaitlík and Martin Saska. Autonomous capture of agile flying objects using UAVs: The MBZIRC 2020 challenge. Robotics and Autonomous Systems 149:103970, March 2022. URLPDF, DOIBibTeX
@article{vrba_ras2022,
title = "{Autonomous capture of agile flying objects using UAVs: The MBZIRC 2020 challenge}",
journal = "Robotics and Autonomous Systems",
volume = 149,
pages = 103970,
year = 2022,
month = "March",
issn = "0921-8890",
doi = "https://doi.org/10.1016/j.robot.2021.103970",
url = "https://www.sciencedirect.com/science/article/pii/S0921889021002396",
author = "Matouš Vrba and Yurii Stasinchuk and Tomáš Báča and Vojtěch Spurný and Matěj Petrlík and Daniel Heřt and David Žaitlík and Martin Saska",
keywords = "Unmanned aerial systems, Machine perception, Mobile robotics, Aerial safety",
pdf = "data/papers/vrba_ras2022.pdf"
}
M Vrba and M Saska. Marker-Less Micro Aerial Vehicle Detection and Localization Using Convolutional Neural Networks. IEEE Robotics and Automation Letters 5(2):2459-2466, April 2020. PDF, DOIBibTeX
@article{vrba_ral2020,
author = "M. {Vrba} and M. {Saska}",
journal = "IEEE Robotics and Automation Letters",
title = "Marker-Less Micro Aerial Vehicle Detection and Localization Using Convolutional Neural Networks",
year = 2020,
volume = 5,
number = 2,
pages = "2459-2466",
doi = "10.1109/LRA.2020.2972819",
issn = "2377-3766",
month = "April",
pdf = "data/papers/vrba_ral2020.pdf"
}
M Vrba, D Heřt and M Saska. Onboard Marker-Less Detection and Localization of Non-Cooperating Drones for Their Safe Interception by an Autonomous Aerial System. IEEE Robotics and Automation Letters 4(4):3402-3409, October 2019. PDF, DOIBibTeX
@article{vrba_ral2019,
author = "M. {Vrba} and D. {Heřt} and M. {Saska}",
journal = "IEEE Robotics and Automation Letters",
title = "Onboard Marker-Less Detection and Localization of Non-Cooperating Drones for Their Safe Interception by an Autonomous Aerial System",
year = 2019,
volume = 4,
number = 4,
pages = "3402-3409",
keywords = "Drones;Three-dimensional displays;Cameras;Target tracking;Robot vision systems;Trajectory;Aerial systems: perception and autonomy;field robots;multi-robot systems;recognition",
doi = "10.1109/LRA.2019.2927130",
issn = "2377-3766",
month = "Oct",
pdf = "data/papers/vrba_ral2019.pdf"
}
