Unmanned Aircraft Positioning System Based on Multi-Source Data Fusion
Qin Tang, Zuyan Wang · 2024
In recent years, the rapid development of UAV and sensor industry technology has led to the widespread use of UAVs in various industries. However, there are significant limitations in the ability to carry only a single sensor for positioning, and it is challenging to ensure the accurate positioning and stable flight of UAVs in complex environments. To address this issue, this design proposes a multi-source data fusion positioning algorithm using a quadrotor UAV as a platform. The algorithm employs the extended Kalman filter (EKF) to integrate the inertial measurement unit (IMU) and binocular camera odometry data to derive precise three-dimensional attitude information. Concurrently, it combines with the Gmapping algorithm to integrate the two-dimensional laser point cloud data to achieve the accurate positioning of the quadcopter UAV. Experimental results demonstrate that the algorithm enables the UAV to achieve accurate positioning indoors while simultaneously meeting the functional requirements of one-key take-off and precise hovering of the quadcopter UAV. This method effectively overcomes the limitations of a single sensor for UAVs and provides a feasible solution for its application in complex environments.