Positioning of UAVs in Urban Environments Using Fusion of TDOA and AOA Data Based on Extended Kalman Filter

Unmanned aerial vehicles (UAVs) have been extensively deployed across a range of applications thank to their flexibility and low cost. While this expansion has significantly improved their operational efficiency and service capacity, it has also posed challenges for UAV supervision and management systems. To address these issues, this paper proposes a three-dimensional (3D) localization method that integrates time difference of arrival (TDOA) and angle of arrival (AOA) measurements based on the extended Kalman filter (EKF). Specifically, for AOA-based positioning, a uniform circular array (UCA) is employed to capture spatial signal characteristics, and the multiple-signal classification (MUSIC) algorithm is applied to precisely estimate the azimuth and elevation angles of incoming signals. In TDOA-based localization, a multipath signal separation and identification algorithm is implemented to enhance robustness against multipath propagation in complex environments. Subsequently, the TDOA and AOA measurements are fused using the EKF, where nonlinear measurement models are linearized via Jacobian matrices to improve computational efficiency and estimation accuracy. Finally, simulation results demonstrate that the proposed hybrid localization method outperforms existing positioning methods that rely solely on AOA or TDOA, achieving a positioning accuracy of approximately 5 m and an angular error within 3°, which is suitable for applications in multipath environments such as urban areas.