D.A. Yudin
Belgorod State Technological University named after V.G. Shoukhov (BSTU after V.G. Shukhov), Engineer, ul. Kostyukova, 46, Belgorod, Russia, tel.: +7(920)200-73-95, This email address is being protected from spambots. You need JavaScript enabled to view it.
V.V. Procenko
BSTU after V.G. Shukhov, Student, ul. Kostyukova, 46, Belgorod, Russia, tel.: +7(4722)30-99-01, This email address is being protected from spambots. You need JavaScript enabled to view it.
G.G. Postolsky
BSTU after V.G. Shukhov, Post-graduate student, ul. Kostyukova, 46, Belgorod, Russia, tel.: +7(4722)30-99-01, This email address is being protected from spambots. You need JavaScript enabled to view it.
A.S. Kizhuk
Doctor of Technical Sciences, BSTU after V.G. Shukhov, Professor, ul. Kostyukova, 46, Belgorod, Russia, tel.: +7(4722)23-05-09, This email address is being protected from spambots. You need JavaScript enabled to view it.
V.Z. Magergut
Doctor of Technical Sciences, BSTU after V.G. Shukhov, Deputy director for research in institute, ul. Kostyukova, 46, Belgorod, Russia, tel.: +7(960)625-44-78, This email address is being protected from spambots. You need JavaScript enabled to view it.
Abstract
While developing autonomous mobile robots an important task is the automatic orientation, localization, positioning and navigation within known and especially unknown scene of observation.
There has been developed computer vision system (CVS), which is able to identify artificial landmarks on images received from video camera with pan-tilt mechanism, and provide calculating robot's deviation from the course set. Robot's route boundary lines and tags in the form of two-dimensional barcodes are considered as such artificial landmarks. On the basis of available dimensional model (with landmarks) it becomes possible for CVS to set the position and orientation of mobile robot. The CVS hardware-software complex includes one camera with pan-tilt mechanism and the onboard computer. The software realized allows searching artificial landmarks in the environments and identifying them at the image. This software also provides control of the drive wheels of the robot in order to eliminate the course deviation calculated on the basis of information on artificial landmarks.
To detect tags in the form of two-dimensional barcodes as the landmarks, the algorithm based on brightness characteristics analysis of image cross-sections has been implemented. To recognize robot's route boundary lines as landmarks, the algorithm based on operations of threshold image transformations, morphological expansion and Hough transform, has been used.
The proposed CVS has been tested on mobile robot with differential drive at BSTU n.a. V.G. Shukhov «Technical Cybernetics» department where special visual landmarks has been used as the target objects.
The proposed computer vision system for automatic positioning and orientation of the robot is cheap and can be used as part of control systems of warehouse automated guided vehicles, service robots, security robots, mobile robots operating in hostile and dangerous for human health environments.
The work has been performed under the Grant of Russian Foundation Basic Research № 12-07-97526-r_tsentr_a «Information and computing intelligent control system for robotic vehicles to solve logistical tasks of industrial and agricultural production», state task of Russian Ministry of Education № 01201256918.
Key words
Robot, vision, moving, object, method, navigation.
Year 2014 Issue number 1 Consecutive issue number 2 Pages 70-75
UDC identifier: 004.93:629.051:621.865.8-5
