Autonomous 3D UAV Localization using Taylor Series linearized TDOA-based approach with Machine Learning Algorithms

Time difference of arrival (TDOA) is the prominent technology for autonomous and real-time three-dimensional (3D) location estimation of the Unmanned aerial vehicles (UAVs). Conventional TDOA localization techniques suffer from the nonlinear optimization problem and hyperbolic intersection to predict the 3D location of UAVs precisely. Therefore, this paper proposes a new positioning Taylor series linearized TDOA-based approach to estimate a precise 3D position of the UAV s. The proposed approach determines the 3D location of the UAV s by evaluating the synchronized difference in arrival time of the signal at spatially separated various anchors nodes. Moreover, machine learning algorithms are applied to optimize the performance of the Taylor series linearized TDOA-based approach and provide a localization solution with comparable accuracy in real-time applications. Consequently, the simulation results are expressed in terms of root mean square errors compared with various machine learning algorithms.