A camera-based indoor position recognition apparatus includes: a floor capturing unit capturing a floor image by controlling a camera installed on a ceiling of an indoor space; a grid map generator constructing a grid map including cell images and position values by dividing the captured floor image into cells of a predetermined size and assigning the position values to the cells; a transceiver performing a data communication with a robot in the indoor space; and a controller providing an identifier-registration and a current position-information to the robot in accordance with an identifier-registration-requesting message and a position-information-requesting message of the robot, wherein the identifier-registration-requesting message and the position-information-requesting message of the robot are delivered from the transceiver.