Object tracking system and method capable of stable position estimation for object whose shape is recognized unstably by radar sensor

A system comprising:

• at least one autonomous driving sensor comprising at least one radar sensor;
• at least one processor;
• and
• at least one memory storing computer program instructions that, based on being executed by the at least one processor, perform artificial intelligence (AI)-based operations of recognizing and estimating a position of an object in a surrounding environment based on detection point data obtained from the at least one radar sensor, the operations comprising:
• performing preprocessing on the detection point data obtained from the at least one radar sensor, to generate preprocessed data;
• performing clustering on the preprocessed data to generate at least one cluster of points;
• generating cluster information from the at least one cluster of points, wherein the cluster information comprises position information and a velocity vector;
• generating a prediction of a track that has been updated up to a previous time;
• performing association between the cluster information and the prediction of the track that has been updated up to the previous time;
• and
• generating an updated estimate of the track for a current time, based on the cluster information and the prediction of the track that has been updated up to the previous time,
• wherein generating the updated estimate of the track comprises:
• generating updated horizontal position information by applying a moving average to horizontal position information among the position information in the cluster information;
• and
• updating track information, which includes heading information of the track, by recalculating a horizontal velocity of the velocity vector based on the position information including the updated horizontal position information.