Visual Feedback for Control Using Haar-Like Classifier to Identify the Quadcopter Position

In order to precise control of unmanned aerial vehicles (UAV), there is a need to get information about UAV's attitude. In the paper a visual feedback is used in order to determine an altitude of the quadcopter. In a case of measure in range of meters or hundreds of meters, typical methods based on barometric sensor, or radar altimeter are used. However, in a case of range of centimeters or single meters those methods may fail. Visual feedback allows control in range of centimeters or decimeters, depend of setup of visual control. The Haar-like classifier was created in order to find in a streaming video the quadcopter image. Experimental verification of the visual feedback for control of a quadcopter is presented.