Construction of an Environmental Map including Road Surface Classification Based on a Coaxial Two-Wheeled Robot

  • Minoru Sasaki Mechanical Engineerng Department, Gifu University, Japan
  • Eita Kunii Mechanical Engineerng Department, Gifu University, Japan
  • Kojiro Matsushita Mechanical Engineerng Department, Gifu University, Japan
  • Joseph K. Muguro Mechanical Engineerng Department, Gifu University, Japan

Abstract

This study details the construction of an environmental map as well as road surface identification based on a coaxial two-wheel robot. Two-wheeled robots are faced with the challenge of oscillatory motion making it hard to get accurate sensor data which is needed in mapping and localization. The proposed system utilizes Robot Operating System (ROS) for data fusion and a modified Simultaneous Localization and Mapping (SLAM) algorithm to reduce oscillatory motions as well as generate environmental map in real-time. Deep learning was used to perform image segmentation to classify road surface. From the results, posture control was verified as well as generation of indoor/outdoor environment maps from sensor data and image processing. The study reported a reduction of oscillatory motion from 40 to 10 degrees. Image segmentation reported a prediction confidence of 80% or more which was adequate for map generation.
Published
Mar 11, 2020
How to Cite
SASAKI, Minoru et al. Construction of an Environmental Map including Road Surface Classification Based on a Coaxial Two-Wheeled Robot. JOURNAL OF SUSTAINABLE RESEARCH IN ENGINEERING, [S.l.], v. 5, n. 3, p. 159-169, mar. 2020. ISSN 2409-1243. Available at: <http://sri.jkuat.ac.ke/ojs/index.php/sri/article/view/801>. Date accessed: 05 july 2020.

Most read articles by the same author(s)

Obs.: This plugin requires at least one statistics/report plug-in to be enabled. If your statistics plugins provide more than one metric then please also select a main metric on the admin's site settings page and/or on the journal manager's settings pages.