Enhanced adaptability of UWB/INS for unmanned surface vehicles in GPS-denied environments: Ranging error identification and compensation

The navigation performance of global positioning system (GPS) and inertial navigation systems (INS) integrated navigation systems is severely degraded under bridges or in harbors, greatly affecting the navigation safety of unmanned surface vehicles (USVs). Ultra-wideband (UWB) ranging technology is a competitive alternative to GPS. However, it is significantly affected by non-line-of-sight (NLOS), limited diffraction line-of-sight (LD-LOS), and signal interruption errors. Therefore, this paper mainly designs a positioning estimation based on the model and an event-triggered mechanism to identify and compensate for ranging errors of the UWB to improve navigation accuracy in GPS-denied environments. First, the USV models are used to estimate the current position. The distance between the estimated position and the UWB base stations is a reference for calculating ranging errors. An event-triggered mechanism is introduced, and an error threshold is set as the trigger signal. Outliers are compensated, and their associated trust weights are adjusted. Finally, the confidence weights provided by the event-triggered mechanism are utilized in the extended Kalman filter (EKF). The advantages of the proposed algorithm over EKF and UWB are experimentally verified, and the positioning errors are reduced by 42.6 % and 69.2 %, respectively. • Reduces USV positioning error by 42.6% vs EKF and 69.2% vs standalone UWB, enhancing safety in GNSS-denied areas. • Employs model-based estimation to compensate for UWB errors, enhancing robustness in UWB/INS navigation. • An adaptive weighting mechanism optimizes fusion reliability for improved positioning accuracy in dynamic conditions.