Interest in hovering unmanned aerial vehicles (hUAV) has increased in recent years.  Hovering is made possible because these vehicles allow for more degrees of freedom in movement as compared to traditional fixed wing UAVs.  Hovering allows the vehicle to remain in place when needed, fly closer to objects of concern, and maneuver in ways that other UAVs cannot.  Hovering unmanned vehicles have been proposed for uses in crop dusting, remote sensing, cinematography, aerial mapping, tracking, inspection, law enforcement, surveillance, search and rescue, etc. 

     Current autonomous micro-UAVs require a computer on the ground to process video because equipment to do so cannot be carried on the UAV itself.  This limits the range of the aircraft and hinders its ability to perform certain tasks because of the time required to transmit images and commands back and forth from the ground station.

     A useful micro-UAV with an onboard vision system needs to be rugged enough to handle a military environment, run in real-time, be compact, require a minimum amount of power, and be standardized and flexible enough to work on a variety of existing and future platforms.  The system must be able to follow general guidance, as opposed to needing detailed commands, from the operator.  The system must be able to detect and avoid obvious obstacles.  The system must also have minimal electro-magnetic signature so as to be less likely to be detected and less likely to interfere with other systems.

     This research looks at the autonomous micro-hUAV being developed at the Robotic Vision Lab (RVL) at Brigham Young University (BYU).  Specifically we focus on the four-rotor micro-helicopter platform and the embedded vision system design. Our research focus is not in developing the underlying helicopter technology or control, so much as it is in developing a vision system to be used primarily on a hUAV and on other space or weight restrained applications.

 Project Sponsors:

 Collaborators:

 Graduate Students:

 Aaron Dennis, Spencer Fowers, Kirt Lillywhite, and Beau Tippetts

1. D.J. Lee, K.D. Lillywhite, S.G. Fowers, B.E. Nelson, and J.K. Archibald, “An Embedded Vision System for an Unmanned Four-rotor Helicopter”, SPIE Optics East, Intelligent Robots and Computer Vision XXIV: Algorithms, Techniques, and Active Vision, vol. 6382-24, 63840G, Boston, MA, USA, October 1-4, 2006.