METHOD AND SYSTEM FOR UNSUPERVISED LEARNING OF ROAD SIGNS USING VEHICLE SENSOR DATA AND MAP DATA

A method, system, and computer program product is provided, for example, for identifying a type of road sign. The method may include receiving a plurality of observations for the road sign. The method may also include clustering the plurality of observations based on at least a location and a heading of the road sign. Further, the method may include filtering the plurality of observations in the cluster based on a speed value and a lateral offset of each of the plurality of observations. Also, the method may include identifying the type of the road sign based on the filtering.