METHOD FOR ESTIMATING LOCATION BASED ON OBJECT RECOGNITION USING KALMAN FILTER

摘要

Disclosed is a method of self-localization of a moving body using a Kalman filter in a vision system based automatic navigation technology. The method of self-localization of the moving body based on an object recognition using the Kalman filter according to an aspect of the present invention comprises the steps of: acquiring images using a camera provided in the moving body; recognizing objects in the images acquired and extracting location information of the moving body based on a coordinate information of the recognized objects; and estimating the location of the moving body by applying the location information to the Kalman filter. [Reference numerals] (110) Camera; (120) Map storage unit; (130) Object recognition unit; (140) First calculation unit; (150) Second calculation unit; (160) Location estimating unit; (AA) One object coordinate; (BB) Two or more object coordinates