Augmented EKF localization for mobile robots in urban environments

Abstract

We propose the augmented EKF Localization for mobile robots in urban environments. Odometer, GPS (Global Positioning System), gyroscope, and camera are used as the localization sensors. The proposed method is aimed for the mobile robot working in urban environments where the traffic lanes exist. Through a camera, the robot measures the lateral offset from the robot position to the centerline of road. Furthermore, the lateral offset is included probabilistically as robot's augmented state to strengthen the robot position estimate. The simulation shows that our proposed method has superior performance compared to the other EKF versions.