2020 | vol. 68 | nr. 2 | art. 12

Agricultural Machinery Movement Navigation System based on Binocular Vision Detection Technology

Chengcheng Wang, Yaqiu Liu, Peiyu Wang
The traditional field test method for agricultural machinery navigation system is constrained by the growth state of the crop. The problem of missing the appropriate period of crop growth will directly lead to the extension of the development cycle and the increase of the cost. In this paper, a new method of binocular vision navigation test of agricultural machinery based on virtual reality technology is put forward. The method takes agricultural machinery as the working machine, and the seedling stage cotton is the target crop. In the virtual reality environment, the three-dimensional geometric model of the field crop row scene is set up. The field path which is identified by the binocular vision method is the target path. According to the relative position relationship between the current driving path and the target path of the agricultural machinery, the model steering wheel angle is calculated and controlled before the agricultural machinery. The results show that the virtual experiment system designed in this paper can carry out the navigation experiment based on binocular vision in the field environment of the field, which can provide theoretical basis and experimental data for the test and improvement of the navigation control system.
Keywords: Binocular Vision Detection, Agricultural machinery movement, Motion tracking, Navigation
To cite this article: Chengcheng Wang, Yaqiu Liu, Peiyu Wang. “Agricultural machinery movement navigation system based on binocular vision detection technology” in Electrotehnica, Electronica, Automatica (EEA), 2020, vol. 68, no. 2, pp. 112-118, ISSN 1582-5175.