Methodology for Localizing an Indoor Drone Using Onboard Sensors and Validating by a Fixed Lidar

This work presents a method for drone localization in indoor environments, which poses a significant issue due to the unreliability or lack of GNSS signals. Drone positional information is crucial in indoor surveillance applications. So, to address this issue, this paper proposes a method for drone localization that uses onboard sensors (camera and IMUs) and ArUco markers placed in the environment. The positions estimated from Visual Inertial Odometry (VIO) are combined with the ones from marker detection through a Kalman Filter (KF). Then, an external fixed Lidar sensor is employed to track the drone. This latter localization serves as a ground-truth for the former one, and for possible successive elaborations of the positional data. The two estimations are aligned and synchronized. The methodology is evaluated using a DJI FLY 2S drone and an Ouster OS1 Lidar in a closed environment.