Simulation of Autonomous Precision Landing for Multirotor Drones

Lee, Wen Kit (2021) Simulation of Autonomous Precision Landing for Multirotor Drones. Final Year Project (Bachelor), Tunku Abdul Rahman University College.

Text Lee Wen Kit.pdf Restricted to Registered users only Download (5MB)

Abstract

In current era, drones are widely used in many sectors. For example, agriculture, surveillance system, delivery system, building etc. The autonomous landing is essential for the drone as it can reduce manpower in landing the drone. Most of the drones in the market mainly uses global positioning system (GPS) only for autonomous landing. Since GPS doesn’t show the exact position, the landing of the drone is not precise which is dangerous. Therefore, this thesis provides an autonomous precision landing solution for drone at fixed location. Moreover, full software development is done in this project. This means that there is no hardware involved for this project. This drone and landing pad involved in this project is virtually created. Besides, visual-based techniques are used in the autonomous precision landing solution. This is due to most of the drones in the market consists of a camera. Or else, a normal camera can be installed on the drone with low cost. Landing pad is one of the main parts for the visual-based autonomous precision landing system. This is because it is required to be detected by the drone during autonomous precision landing. In the project, ArUco markers are used for landing pad design as it allows fast and reliable detection. The program for autonomous precision landing is developed and validated in order to provide a solution that enable drone to land on the landing pad safely and precisely with less than 50 cm error. Lastly, Design D of landing pad produces the best overall result for precision landing with landing error of 0.1048 m at windless condition and 0.1476 m with wind condition.