Summary: For solving the problem of low efficiency and low localization accuracy about EKF-SLAM algorithm in large scale and complicated environment, an improved method by adding an adaptive observation range to EKF-SLAM algorithm was presented. By setting the constraints, the observed environment scale and the number of observed landmarks for mobile robot were constrained. Deleting the landmarks which has observed but beyond the constrained scale by judge the relationship between the observation range and maximum distance, dynamically adjusting the observation range by comparing the number of observed landmarks and number of landmarks. Simulation results demonstrate that the proposed method could reduce the computational and improve efficiency and localization accuracy.

Key words: mobile robot, improved EKF-SLAM algorithm, landmarks.