High precision position estimation method through road shape classification-based map matching and autonomous vehicle thereof

An autonomous vehicle comprising:

• a sensor configured to:
• generate lane image information for a road;
• and
• detect a vehicle speed and a yaw rate as sensor measurement;
• a precision map storage having a road map of an autonomous traveling precision map;
• and
• a high precision position estimation system configured to calculate and output at least one of latitude, longitude, or traveling direction of a vehicle,
• wherein the high precision position estimation system comprises:
• a controller configured to:
• classify a lane into a straight lane, a curved lane, and a clothoid curve lane by the lane matching a sensor lane segment of the lane image information with a map lane segment of the autonomous traveling precision map;
• generate the movement amount calculation values of geometric feature-based map matching for the straight lane and the curved lane;
• and
• generate the movement amount calculation value of an optimization-based map matching for the clothoid curve lane as a covariance matrix;
• and a vehicle state estimator configured to:
• perform the estimation position prediction by combining the sensor measurement value with the covariance matrix by an extended Kalman filter;
• and
• calculate and output the latitude, the longitude, and the traveling direction by the estimation position prediction.