Title :
Multiple robot formation keeping and cooperative localization by panoramic view
Author :
Kang, Hyun-Deok ; Kawk, Jae-Hyuk ; Kim, Chang-Hwan ; Jo, Kang-Hyun
Author_Institution :
Korea Inst. of Sci. & Technol., Seoul, South Korea
Abstract :
This paper describes the method of formation keeping and cooperative localization of multiple robots with panoramic view. It is to estimate a location of robot in panoramic image with 360 field of view horizontally and then to compute the relative distance between the robots by triangulation. This method is able to control predefined formation when a robot extracts other robot and landmark in panoramic view and also estimates the geometric correspondence between robots and landmarks. Moreover, proposed method is independent with a displacement sensor such as GPS, inertial and so. If the sensory information such as GPS or inertial data is interrupted which the robot moves into a tunnel or forest and so. A panoramic imaging data is used to control in multiple robots alternatively. Therefore, robot computes his location and their formation which cannot handle the GPS or inertial sensor. The angle of between robots in panoramic image is the location as an extracted feature in panoramic image and the angle of predefined formation of robots is also estimated. Each robot observes the other robots using their panoramic view and determines the angle to relative to viewing angle in other robot´s view. A proposed method is verified in synthetic environment and we showed the effectiveness about control and keeping the formation of multiple robots in outdoor environment even when pose and angular sensors are not working.
Keywords :
computational geometry; feature extraction; mobile robots; multi-robot systems; path planning; robot vision; stereo image processing; GPS; angular sensor; cooperative localization; displacement sensor; feature extraction; inertial sensor; multiple robot formation keeping method; outdoor environment; panoramic imaging view; panoramic stereo geometric correspondence; pose sensor; robot location estimation; synthetic environment; Cameras; Computational geometry; Data mining; Force control; Global Positioning System; Mobile robots; Optical imaging; Robot control; Robot sensing systems; Robot vision systems; Cooperative localization; Formation keeping; Panoramic view;
Conference_Titel :
ICCAS-SICE, 2009
Conference_Location :
Fukuoka
Print_ISBN :
978-4-907764-34-0
Electronic_ISBN :
978-4-907764-33-3