UWB and IMU Fusion Based on Kalman Filter in Mobile Robot Localization System

Abstract

Nowadays, mobile robots are used in a wide variety of fields, such as manufacturing, agriculture, space and underwater exploration, and healthcare. The localization process is crucial to mobile robots and refers mainly to the precise determination of the coordinates where the system is present at a certain moment. Ultra-wideband (UWB) and inertial measurement unit (IMU) are commonly used techniques in localization systems. However, UWB can’t avoid Non-line-of-sight (NLOS) propagation errors due to obstacles and IMU always suffers from error accumulation over time. In my research, IMU and UWB are combined optimally in the Kalman filter to overcome their limitations in specific situations. A new method to detect UWB measurement anomaly is proposed based on the difference between the velocity calculated by UWB measurement and IMU data. The complementary filter is used to combine the accelerometer data and gyroscope data to derive the roll and pitch degree. Kalman filter parameters are adjusted in different situations to help the localization system perform better and provide reliable position information to the control system to complete the tracking task.