|
|
Design of quadcopter contol sytem with visual guidance
Pokorný, Ondřej ; Věchet, Stanislav (oponent) ; Andrš, Ondřej (vedoucí práce)
This thesis deals with the design and comprehensive implementation of a quadcopter control system with visual guidance towards a printed marker. The system consists exclusively of low-cost, commercially available hardware and open-source or custom software. The units used are, for example, microcomputer Raspberry Pi 3 Model B, microcontroller Arduino Nano, flight controller Omnibus F3, etc. In the first part, the structure of the system is outlined and the properties and functions of the components described. Following is an overview of the communications used and their versions specific to flying platforms. Finally, the architecture of the custom software is described together with the inner workings of the single parts and the reasons for their presence in the code. The second part details the use of the ArUco augmented reality library for pose estimation, including the measures introduced to compensate for the inherent flaws of this system. This part also contains a description of the control algorithm development and of the subsequent testing of the implemented solution, as well as suggested further steps.
|
|
|
Design of quadcopter contol sytem with visual guidance
Pokorný, Ondřej ; Věchet, Stanislav (oponent) ; Andrš, Ondřej (vedoucí práce)
This thesis deals with the design and comprehensive implementation of a quadcopter control system with visual guidance towards a printed marker. The system consists exclusively of low-cost, commercially available hardware and open-source or custom software. The units used are, for example, microcomputer Raspberry Pi 3 Model B, microcontroller Arduino Nano, flight controller Omnibus F3, etc. In the first part, the structure of the system is outlined and the properties and functions of the components described. Following is an overview of the communications used and their versions specific to flying platforms. Finally, the architecture of the custom software is described together with the inner workings of the single parts and the reasons for their presence in the code. The second part details the use of the ArUco augmented reality library for pose estimation, including the measures introduced to compensate for the inherent flaws of this system. This part also contains a description of the control algorithm development and of the subsequent testing of the implemented solution, as well as suggested further steps.
|
host ::
RSS.