Fusion positioning method with UWB/IMU/odometer based on the improved UKF

A multi-sensor fusion positioning method based on the improved unscented Kalman filter (UKF) algorithm is proposed to address the problem that ultra wide band (UWB) cannot be precisely located in the environment with unknown statistical properties with large noise variations. The information of UWB, IMU and odometer encoder is fused centrally. An adaptive factor is introduced to update the measurement noise covariance matrix in real time to update the observation noise, and the fading factor is added to suppress the filtering divergence to achieve an improvement for the traditional UKF algorithm. Based on the two-wheel differential robot and UWB hardware platform, the proposed fusion positioning method is evaluated by simulation and field experiment in UWB line-of-sight (LOS) and non-line-of-sight (NLOS) scenarios respectively, and compared with the single-sensor positioning including UWB, IMU, and fusion algorithms based on the extended Kalman filter (EKF) and UKF. The results show that the proposed fusion positioning method can maintain high accuracy in both LOS and NLOS scenarios, and the improved UKF algorithm improves the positioning accuracy by about 22.8% and 13.1% over EKF and UKF algorithms respectively in NLOS scenarios. © 2023 Editorial Department of Journal of Chinese Inertial Technology. All rights reserved.