Development of Boundary Detection Methods Based on Images for Path Following of Autonomous Tractor 


Vol. 46,  No. 11, pp. 2078-2087, Nov.  2021
10.7840/kics.2021.46.11.2078


PDF
  Abstract

This paper proposes the autonomous traveling methodology of tractors based on an image recognition unlike existing GNSS-based tractor studies. Among the branches of LaneNet, which is a CNN-based lane detection model, we have developed a model that can be used in special agricultural work environments such as paddy or dry fields by modifying it to use only a part considering to the purpose. Image datasets for tillage/untilled boundaries in the fields required for learning were collected directly, and the working boundaries between tilled/untilled areas using the learned model were detected. The steering angle to keep the distance between the boundary and the reference line was calculated so that the tractor can autonomously travel using the detected boundary line. The calculated steering angle was transmitted to the tractor in real time through CAN communication and used it for steering control of the tractor. As a result of performing steering control for the straight section based on the proposed model, it was confirmed that autonomous traveling was possible with an error of up to ±9.51cm.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

D. Seo, J. Won, C. Yang, G. Kim, K. Kwon, K. Kim, Y. Hong, H. Ryu, "Development of Boundary Detection Methods Based on Images for Path Following of Autonomous Tractor," The Journal of Korean Institute of Communications and Information Sciences, vol. 46, no. 11, pp. 2078-2087, 2021. DOI: 10.7840/kics.2021.46.11.2078.

[ACM Style]

Dasom Seo, Jin-Ho Won, Changju Yang, Gookhwan Kim, Kyung-Do Kwon, Kyoung-Chul Kim, Youngki Hong, and Hee-Seok Ryu. 2021. Development of Boundary Detection Methods Based on Images for Path Following of Autonomous Tractor. The Journal of Korean Institute of Communications and Information Sciences, 46, 11, (2021), 2078-2087. DOI: 10.7840/kics.2021.46.11.2078.

[KICS Style]

Dasom Seo, Jin-Ho Won, Changju Yang, Gookhwan Kim, Kyung-Do Kwon, Kyoung-Chul Kim, Youngki Hong, Hee-Seok Ryu, "Development of Boundary Detection Methods Based on Images for Path Following of Autonomous Tractor," The Journal of Korean Institute of Communications and Information Sciences, vol. 46, no. 11, pp. 2078-2087, 11. 2021. (https://doi.org/10.7840/kics.2021.46.11.2078)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880