Title :
Landmark detection of in-pipe working robot using line-laser beam projection
Author :
Lee, Dong-Hyuk ; Moon, Hyungpil ; Choi, Hyouk Ryeol
Author_Institution :
Dept. of Mech. Eng., Sungkyunkwan Univ., Suwon, South Korea
Abstract :
In-pipe works such as inspection or cleaning which are difficult to be done by human seem to be suitable tasks for automatic robots. Since the wireless communication is hardly accessible in pipeline and the wired communication severely limits the mobility of in-pipe robot, the in-pipe working robot should work autonomously and independently. Landmark based localization which is widely used in mobile robot also can be applied to localization of in-pipe working robot in pipeline environment. In a pipeline, specific structures such as branches or elbows can be treated as landmarks which are distinguishable from the environment. In this paper, we introduce a noise tolerant landmark detection method using line laser beam projected on internal surface of the pipeline. Once the line laser is projected on the surface of landmarks such as elbow or branch, a unique line pattern is generated. By using this line pattern the robot can obtain the information to perform the localization. Line laser beam used in this method has distinct color like red or green which is easily noticeable from real environment in pipeline. Therefore the image processing procedure to observe this line becomes easier even in rusted and contaminated environments.
Keywords :
SLAM (robots); cleaning; industrial robots; inspection; mobile robots; object detection; pipelines; robot vision; automatic robot; branches; cleaning; elbows; inpipe working robot; inspection; landmark based localization; landmark surface; line laser beam projection; line-laser beam projection; mobile robot; noise tolerant landmark detection method; pipeline environment; wireless communication; Cameras; Elbow; Inspection; Laser beams; Lasers; Pipelines; Robots; In-pipe robot; Landmark; Pipe inspection;
Conference_Titel :
Control Automation and Systems (ICCAS), 2010 International Conference on
Conference_Location :
Gyeonggi-do
Print_ISBN :
978-1-4244-7453-0
Electronic_ISBN :
978-89-93215-02-1