Autonomous mobile robots are used in warehouse logistics and automatic unmanned production. This paper presents an algorithm for sensor fusion of measurements from a UWB-based radio navigation system with encoder measurements. The scientific novelty of the work lies in the use of a UWB-based navigation system to correct the estimates of the mobile robot’s position obtained through dead reckoning using encoder data, which accumulates error over time. The study focuses on a robot with a differential drive. The presented algorithm is based on the Extended Kalman Filter (EKF), where the estimated parameters include the robot’s two-dimensional coordinates, heading in the local coordinate system, linear and angular velocities. The observations consist of estimates of the robot’s two-dimensional coordinates and the angular velocities of the wheels. The paper presents the results of simulation modeling and experimental studies of the proposed algorithm. These results demonstrated a positive effect when integrating measurements from the radio navigation and odometric systems, resulting in improved accuracy of determining the coordinates, linear velocity, and heading angle.