AMC-LIEKF: An Adaptive Maximum Correntropy Left-Invariant Extended Kalman Filter Method for Tightly Coupled Multi-UWB/IMU Integration

Abstract

The error-state extended Kalman filter (ESKF) is widely used to fuse ultrawideband (UWB) and inertial measurement unit (IMU) data for robust localization in global navigation satellite system (GNSS)-denied environments. However, conventional ESKF-based fusion methods still face challenges such as geometric inconsistency and estimation bias caused by model linearization, as well as sensitivity to non-Gaussian (NG) disturbances. To address these challenges, we propose an adaptive maximum correntropy left-invariant extended Kalman filter (AMC-LIEKF) for tightly coupled (TC) multi-UWB/IMU integration. The proposed method is formulated on a matrix Lie group (LG) to preserve geometric and estimation consistency, which enhances the stability of attitude and position estimation under dynamic motion. Moreover, to further improve robustness against NG disturbances, the maximum correntropy (MC) criterion is incorporated, and we design a dynamic kernel-bandwidth mechanism that adaptively adjusts the UWB measurement weights based on measurement residuals and the received signal strength indicator (RSSI). Field experiments demonstrate that the proposed AMC-LIEKF significantly improves estimation accuracy and stability compared with conventional ESKF- and left-invariant extended Kalman filter (LIEKF)-based methods, particularly in dynamic environments with NG noise.