ISSN 2079-3537      

Scientific Visualization
Issue Year: 2017
Quarter: 2
Volume: 9
Number: 2
Pages: 13 - 25
Authors: B.A. Zalesky (Belarus), V.B. Shuvalov (Russian Federation)
Address: B.A. Zalesky
United Institute of Informatics Problems of National Academy of Sciences, Minsk, Belarus

V.B. Shuvalov
National Research Nuclear University MEPhI (Moscow Engineering Physics Institute), Moscow, Russian Federation
Abstract: The algorithm of autonomous navigation of an unmanned aerial vehicle (UAV) equipped with an autopilot, onboard inertial navigation system, onboard computer and one video camera, preferably mounted on a gyro platform, is proposed. It is designed to return the autonomously flying UAV to the starting point of its route without the use of external navigation signals such as GPS or GLONASS. The idea of the algorithm consists in comparison each of the current frame, made by the onboard camcorder during UAV return home, with frames shot by the camcorder while the apparatus flight from the starting point along a predetermined path.
The offered algorithm can also provide autonomous navigation of the unit without use of external navigation signals throughout the flight. For this it needs a pre-stored video sequence, or a sequence of pictures, for example, satellite imagery, completely covering any realized route of the apparatus from the starting point to the destination point, and suitable for solving the problem of recognition.
Results of computer modeling of the autonomous navigation process and its visualization are presented. Simulation of the UAV flight has been done, first, based on videosequences made by smartphones, and then with help of movies captured by the onboard camcorder of the quadcopter DJI Fantom 3 Professional.
Computational experiments have shown robustness and sufficient accuracy of the offered algorithm.
Language: English

Open Article
Open Article
in Russian translation
   Download ZIP archive
Download ZIP archive
in Russian translation