Cooperative Positioning Algorithm Based on Manifold Gradient Filtering in UAV-WSN

Owing to the flexible and lightweight characteristics, unmanned aerial vehicles (UAVs) are widely used in the field of cooperative navigation and positioning technology. Furthermore, wireless sensor networks (WSNs) provide a bridge for information exchange and fusion between the cooperative UAV nodes. Focusing on the demand of the lightweight and real-time dynamic requirements in the UAV-WSN, this article proposes a cooperative positioning algorithm based on manifold gradient filtering assisted by geometric dilution of precision (GDOP). In this algorithm, the measurement model between the cooperative sensors nodes in UAV-WSN is used to construct the Riemannian manifold. By deriving the gradient on the manifold, the fastest descent direction in iteration can be determined. The GDOP corresponding to the geometric configuration of the cooperative UAV nodes is derived, which is applied to modifying the iterative descent rate and making the algorithm converge quickly. The simulation results show that the algorithm converges fast and has good performance in accuracy. Moreover, the algorithm also exhibits a certain level of robustness in extremely harsh environments.