3D Pure Pursuit Guidance of Drones for Autonomous Precision Landing

Landing a UAV precisely to its target location with challenges like external disturbance and inaccurate GPS localization makes it more difficult. This paper proposes a simplified pure pursuit-based method to approach the target landing location using virtual targets. Firstly the Aruco marker placed on the landing pad is identified using an OpenCV framework which helps the UAV for self-localization. In the next step, drone way-points are generated using the proposed algorithm where the 3D space problem is converted to a 2D cartesian system followed by classical pure-pursuit. This method generates a smooth arc for the Drone to follow and avoid crashing while landing. A PID controller ensures that the Drone follows the generated trajectory within a reliable margin of error. This method delivers an efficient and computationally less expensive landing approach for autonomous drones.