Kolloquium Vortrag
The use of small Unmanned Aerial Vehicles (UAVs) has increased rapidly in many civilian and domestic researches. UAVs are usually equipped with one embedded camera at least, and the visual information is used for navigation and for controlling the vehicle itself.
This presentation is the final presentation for a master thesis that uses the visual information received from an embedded camera attached to a quadrocopter to find a predefined pattern located in a semi-structured outdoor area, and to lead the quadrocopter to a predefined position above this pattern.
The general used control structure was a position visual serving technique, which was essentially built on by estimating the quadrocopter position then building the control laws using this estimation. The position estimation was done depending on a combination of scale-Invariant detection strategy and a camera model-based algorithm. Meanwhile, the control part was designed following LQG design method after linearising the theoretical model.
