Method for identifying objects in a traffic space

Abstract

A method for identifying objects in a traffic space by means of a sensor system arranged at or in a host vehicle comprises the steps of determining the positions of a plurality of object surface spots detected by the sensor system in a vehicle coordinate system, assigning the determined positions of object surface spots to at least one detection group by means of a clustering algorithm, wherein the clustering algorithm uses clustering regions defining threshold distances, and identifying at least one object by determining that positions of object surface spots belonging to the same detection group form part of a single object. The determined positions of object surface spots are transformed in a transformation process and the clustering algorithm is applied to the transformed positions of object surface spots. The transformation process includes a directional scaling step that renders the clustering regions non-circular in the vehicle coordinate system.

Dependencies are inferred from the claim text.