Adaptive Extended Kalman Filter-Based Fusion Approach for High-Precision UAV Positioning in Extremely Confined Environments

For unmanned aerial vehicle (UAV)-based smart inspection in extremely confined environments, it is impossible for precise UAV positioning with global positioning system, owing to the satellite signal block. Therefore, the ultrawideband (UWB)-based technology has attracted extensive attention under such circumstances. However, due to the unpredictable propagation condition and the time-varying operational environment, the localization performance oscillation caused by the changing measurement noise may lead to the instability of UAV. To mitigate the effects, in this article, a high-precision UAV positioning system which integrates the inertial measurement unit and UWB with the adaptive extended Kalman filter (EKF) is proposed. Compared with the traditional EKF-based approach, the estimated and recorded information from previous processes is exploited to adaptively estimate and further control the estimation of the noise covariance matrices for the performance improvement. Finally, simulations and experiments have been conducted in extremely confined environments. According to the results, the proposed algorithm can significantly improve the position update rate, the median positioning error, the 95
${\text{th}}$
percentile positioning error, and the average standard deviation into 88 Hz, 0.102 m, 0.192 m, and 0.052 m, which is applicable for applications in focused environments.