Improving Accuracy of the IMU/UWB Fusion Positioning Approach Utilizing ESEKF and VBUKF for Underground Coal Mining Working Face

The high-precision positioning of the shearer in underground coal mines remains a significant challenge task for realizing automation and intelligent coal mining. Most of the presently conventional positioning approaches, nonetheless, tend to deliver low accuracy and unreliable estimations in the GPS-denied underground coal mine environment. Additionally, the current experiments remain in the simulation verification stage while lacking of the real field experiments. In order to improve reliability and positioning accuracy of target node (TN) in coal mines, this article proposes a fusion positioning scheme with the combination of the inertial measurement unit (IMU) and ultrawideband (UWB) based on error state extended Kalman filter (ESEKF) approach with the utilization of the available measurements obtained from IMU system and the established UWB localization system. Subsequently, the variational Bayesian UKF (VBUKF) method that incorporates the variational Bayesian (VB) and unscented Kalman filter (UKF) is implemented to the result of fusion localization with consideration of the changeable measurement noise to further eliminate the trajectory error and refine the final estimation accuracy. Furthermore, to bridge the research gap of the real field experiments using the IMU/UWB fusion system, the experimental investigation is performed to evaluate the effectiveness of the proposed method in the underground coal mine environment. The experimental results demonstrate that the presented ESEKF method, after applying the VBUKF smoothing, is capable to achieve better performances than the contrasted methods in terms of localization, and also can significantly improve the positioning accuracy of the shearer in the coal mining working face.