Skip to Content

Visually Guiding Autonomous Helicopters for Civilian Tasks

Congresses name: 

UAV 2007 Conference, Workshop on Mini UAV

Place: 

Paris, France

Date: 
June 2007

Abstract:
UAVs (Unmanned Aerial Vehicles) are becoming more popular for civil tasks due to their unique flight capabilities. Unmanned Aerial Systems have been an active area of research in recent years. The use of aerial vehicles in civilian applications for surveillance and inspection continues to grow due in part to the emergence of cheap low power vision systems. Vision for flight control encompasses a broad range of research areas such as object detection and tracking, position estimation, sensor fusion with inertial navigation and GPS, and multivariable non-linear system modeling and control. We believe that computer vision is much more than a technique to sense and recover information from the environment. It should play a main role in the UAVs functionality. Our current research focus lies in the development of techniques that allows UAV to maneuver in spaces using visual information as the main input source. We investigate techniques that allow an UAV to perform maneuvers towards features of interest when GPS signal is not reliable, e.g. signal dropouts (usually in urban areas or when flying through terrestrial urban canyons or when operating on remote planetary bodies). We investigate visual servo control techniques that use velocities of suitable image features parameters to compute the references for the flight control for driving the UAV. The present works explores a research field in computer vision for UAVs in four main active research lines, such as visual servoing, trajectory planning, power line inspection and stereo-based visual navigation.