A Vehicle Localization System Using Visual Road Features from Monocular Camera

This paper proposes a vision-based perception system of using a monocular camera to detect the high-level features in the form of road markers and driving lanes for providing reliable visual measurements. For road marker recognition, the template matching-based method is applied to recognize the marker type and estimate the corresponding position and orientation as well. For driving lane detection, the proposed gradient orientation consistency combined the Inverse Perspective Mapping (IPM) spatial constraints is used to initialize the clearly visible lanes in the initial detection. Finally, the particle filter is employed to integrate the visual measurements, inertial measurement unit (IMU), and GPS complementarily for correcting GPS errors. The proposed vehicle localization system is evaluated in the real driving scenario of the standard campus environment within varying illumination condition.