Addressing the issues of low positioning accuracy or inability to locate in closed spaces by the Global Navigation Positioning System (GNSS), as well as the lack of positioning accuracy and stability in non-line-of-sight environments by the Ultra-wideband Pulse Indoor Positioning Technology (UWB), this paper proposes to integrate UWB positioning technology and Inertial Sensor Technology (IMU) based on the Maximum Correlation Entropy Kalman filter. This approach includes modeling of measurement noise, reducing the weight of abnormal measurements to minimize the impact on state estimations, and combining UWB and IMU measurement data to resolve issues of poor UWB positioning accuracy and positioning result offsets in non-line-of-sight environments. In simulation experiments, multiple base stations were used to locate moving targets. The results demonstrated that the proposed algorithm effectively improved the accuracy and robustness of UWB indoor positioning technology compared to the Kalman filter and Extended Kalman filter.