UWB/IMU integration approach with NLOS identification and mitigation

Getting the location information of a mobile station (MS) is very important in many applications. The position of a MS in a ultra-wideband system can be determined based on the distance measurements between the MS and several base stations (BSs), using the time of arrival (TOA) method. However, the Non-Line-of-Sight (NLOS) cause inaccuracy position estimation. In this paper a novel NLOS identification and accurate range measurements selection approach based on the acceleration measurements is proposed. Additionally, the Iterative Extended Kalman Filter (IEKF) is combined with the proposed approach to effectively reduce the NLOS errors. The position estimation accuracy of the system based on IEKF with and without the proposed selection approach is compared. The performance of an ultra-wideband/inertial measurement unit (UWB/IMU) tightly coupled approach is also used as comparison. The most accurate results are achieved with the IEKF with proposed approach both in simulation and real field tests among all the approaches.