An Adaptive Kalman Filter Target Tracking Algorithm for NLOS Environments

Aiming at the problems of Non-Line-of-Sight (NLOS) propagation and multipath interference faced by UAV positioning and tracking in urban environments, traditional filtering methods tend to suffer from trajectory inaccuracy and insufficient stability in time-varying noise scenarios. To address these issues, this paper proposes an adaptive Kalman filter gain-adjustment-based target tracking algorithm for NLOS environments. By introducing an environmental quality evaluation index and a Mahalanobis distance adjustment factor, the proposed algorithm dynamically tunes the quantitative gain of the filter, which not only ensures high-precision target tracking but also effectively enhances the suppression capability against sudden interference. Simulation results show that in 100 Monte Carlo experiments conducted within a 400m×400m area with occlusion zones, the average abnormal point detection frequency of the proposed algorithm reaches 0.605 Hz, and the mean tracking error is reduced by 28.6%. The results verify the effectiveness of the proposed algorithm in NLOS environments.