A Vision Based Guidance Control Design to Quadrotor Helicopter

• Main Authors: Yu Shan-Jay, 游憲杰
• Other Authors: Yuan- Kai Wang
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/83132707797878357264

碩士 === 輔仁大學 === 電機工程學系碩士班 === 102 === Nowadays, planar vertical take-off and landing (PVTOL) of quadrotors raises the considerable interest in the research topic which covers the integration of the vision systems and the controller design. This thesis presents a system integration structure for the fixed PVTOL. Firstly, the system dynamics with 6-DOF for a quadrotor is investigated. Based on this model, the proportional-derivative (PD) controller associated with the altitude, yaw, roll, and pitch are then derived. Moreover, the fuzzy control design is employed to the flying controllers so that the tracking performance in PVTOL can be improved. By using the invariant moment algorithm, the developed vision system can effectively detect and track the landmark against the variation, rotation, and translation in the image. The coordinate of the gravity of the landmark can be calculated for the visual servoing. To this end, the proposed guidance control system can achieve the vision-based PVTOL of the quadrotor. Our system can carry out the visual target tracking under the take-off, landing, and hovering tasks. Through the simulation and experimental verification, the designed fuzzy controllers can outperform the PD controllers in the visual tracking control performance for the PVTOL of the quadrotor.