A sequential mobile vehicle location method with visual features

Localization is of vital importance to a mobile vehicle system. Most of the existing algorithms are based on laser range finders, sonar sensors, artificial landmarks or GPS information. In this paper, we present a sequential probability location method for mobile vehicle, which uses scale-invariant image features as natural landmarks in unmodified environments. First, we construct a ground truth map with the appearance of the environment in a learning step, then by a proposed sequential matching approach and kd-trees, we could recognize the map and locate the mobile vehicle. In the experiment, we use two cameras, one is left oriented and the other is right. We have try several method, the experiment result shows that the proposed method could locate the vehicle in nearly realtime with higher matching rate than the other approach.