Monocular vision system for construction of workspace map and localization of a mobile robot

Authors: Lazic Bojan
Published in issue: #3(56)/2021
DOI: 10.18698/2541-8009-2021-3-680
Category: Mechanical Engineering and Machine Science | Chapter: Robots, Mechatronics, and Robotic Systems
Keywords: mobile robot, television camera, image special points, distance calculation, map building, monocular vision, robot position
Published: 23.03.2021

The problem of constructing a map of a room using a monocular vision system, i.e., using one television camera, which is located on a mobile robot, is considered. The distance is calculated based on the analysis of the object’s special points displacement on two consecutive images received from the TV camera. Mathematical expressions are given for determining the position of wall points in the global coordinate system. Television cameras of the robot, which are sequentially located at different points of the workspace, take pictures of the walls. Approximation of wall projections is performed by the least squares method. The results of experimental studies of a real room are given, the measurement error is determined.

References

[1] Press H.W., Teukolsky A.S., Vetterling T.W., et al. Numerical recipes. Cambridge University Press, 2007.

[2] Korn G.A., Korn T.M. Mathematical handbook for scientists and engineers. McGraw-Hill, 1968. (Russ. ed.: Spravochnik po matematike. Moscow, Nauka Publ., 1974.)

[3] Miller S.J. The method of least squares. URL: https://web.williams.edu/Mathematics/sjmiller/public_html/BrownClasses/54/handouts/MethodLeastSquares.pdf (accessed: 15.12.2020).

[4] Szeliski R. Computer vision. Springer, 2011.

[5] Patrick F.D. Measurement and data analysis for engineering and science. McGraw–Hill, 2005.

[6] Camera calibration. opencv-python-tutroals.readthedocs.io: website. URL: https://opencv-python-tutroals.readthedocs.io/en/latest/py_tutorials/py_calib3d/py_calibration/py_calibration.html (accessed: 15.12.2020).

[7] Smoothing images. docs.opencv.org: website. URL: https://docs.opencv.org/3.0-beta/doc/py_tutorials/py_imgproc/py_filtering/py_filtering.html#filtering (accessed: 15.12.2020).

[8] Teteryukov D. Prezentatsiya predmeta “Robotics – Forward kinematics” [Presentation of “Robotics – Forward kinematics” course]. Moscow, Skoltekh Publ., 2016 (in Russ.).

[9] Matlab. The language of technical computing. MathWorks, 2005.