Research on UAV Navigation Based on the Kalman Filter and Geomagnetic Matching Algorithm

To address the challenge of low-precision navigation in GPS-denied environments for unmanned aerial vehicles (UAVs), this paper proposes a hybrid geomagnetic matching algorithm based on Ant Colony Optimization (ACO) and Particle Swarm Optimization (PSO), combined with an Unscented Kalman Filter (UKF) for robust dynamic state estimation. The proposed method utilizes ACO for global exploration and PSO for local refinement to enhance matching accuracy between real-time magnetic field measurements and a reference geomagnetic map. The UKF integrates matching results with inertial data to compensate for sensor drift and improve the robustness of the estimation. Simulation results demonstrate that the proposed ACO-PSO-UKF framework significantly enhances positioning accuracy, convergence speed, and resistance to interference compared to traditional methods. The average root mean square error (RMSE) is reduced to 0.54 meters, and the system maintains real-time performance with an iteration time under to 50ms. The algorithm offers a practical solution for autonomous UAV navigation in complex and signal-denied environments, with strong potential for embedded implementation and real-world deployment.