Narrow Gaps Detection and Computing Trajectory for Aggressive Autonomous Objective-Navigating Quadcopter

The proposed paper presents an algorithm for an autonomous quadcopter to perform the act of flying over a narrow gap. Despite some of previous works depended on initial position as well as tracking position sensors for calculating the desired orbit to take the mission, the proposed quadcopter only n...

Full description

Saved in:
Bibliographic Details
Published in2022 6th International Conference on Green Technology and Sustainable Development (GTSD) pp. 454 - 458
Main Authors Le, Ngoc-Hien-Duc, Le, Huy-Phuong, Ta, Tran-Nhat-Minh, Trinh, Tan-Dat, Le, My-Ha
Format Conference Proceeding
LanguageEnglish
Published IEEE 29.07.2022
Subjects
Cameras
central coordinate
generating a trajectory
Geometry
Inertial navigation
Measurement units
narrow gap
Orbits
practical condition
quadcopter
Trajectory
Vision sensors
Online AccessGet full text

Cover

More Information
Summary:The proposed paper presents an algorithm for an autonomous quadcopter to perform the act of flying over a narrow gap. Despite some of previous works depended on initial position as well as tracking position sensors for calculating the desired orbit to take the mission, the proposed quadcopter only needs information coming from inertial measurement unit and vision sensor. The mission for a quadcopter passing through a narrow gap itself, is considered as the challenge problem since both estimating optimal flight path and orienting behavior of quadcopter with respect to non-vertical gap section must be finished at the same time in order to keep fast response in practical condition. We solve this problem by generating a trajectory formed on geometry and current system perception: the proposed quadcopter, equipped with a camera faces toward the gap continuously, determines the central coordinate from the hole, then generates desired flying mission path and repeat the whole process until the quadcopter completely pass through the gap. As a result, we evaluate and demonstrate successfully in many types of conditions, including wind intensity and light disturbance. With testing 10 times in each types, the average result is around 85% and the gap is oriented up to 45 degrees. This is the first work for quadcopter to perform such a mission in real life, only use certain sensor to generate optimal trajectory.
DOI:10.1109/GTSD54989.2022.9989004