Work

Trunk Tracking Control Method and Experiment of Caterpillar Self-Propelled Fruit Collector

Year: 2025

Type: article

Abstract: Highlights RANSAC cylindrical surface fitting algorithm based on PCA principal component analysis achieves high precision extraction of three-dimensional coordinates of fruit tree trunk. An arc curve path planning and control method for switching between ridge direction and trunk collection position is proposed. The proposed trunk tracking control method was experimentally validated in a real orchard environment. Abstract. The forest fruit collector walks along the rows of the orchard and needs to go to the vicinity of the fruit tree trunk and spread the umbrella surface for operation. The traditional collection method mainly relies on manual control of the collector in the vicinity of the fruit tree, which is time-consuming and laborious. To solve the problem of self-tracking of the trunk by the forest fruit collector in the orchard environment in the hilly area, a method of trunk tracking by the collector was proposed. By extracting three-dimensional coordinate information of the fruit tree trunk, effective tracking coordinates are obtained, and combined with the path planning algorithm, tracking control is realized. In this article, the motion characteristics of the collector are analyzed firstly, and the tracking algorithm of pre-sight points is adopted to realize the tracking of orchard trunks. Finally, the algorithm is integrated to construct a complete tracking control system; the crawler tree trunk automatic tracking control in the orchard environment is realized. Experiments show that the maximum offset of error is 0.19 m, RMSE (root mean square error) is 0.027 m, the maximum deflection angle error is 23.7°, and RMSE is 3.6°. In the whole tracking operation process, the collector can travel along the planned path and maintain a stable and reliable operating state. Keywords: Fruit collector, Track, Tracking control, Trunk, Vision.

Source: Applied Engineering in Agriculture

DOI: 10.13031/aea.16168

Authors: Zenghong Ma, Jiawen Yue, Cheng Yin, Xiaoqiang Du, Lijun Zhao, Hongyu Hu

Institutions: Zhejiang Sci-Tech University, Zhejiang University of Science and Technology, Ministry of Agriculture and Rural Affairs, Chongqing University of Arts and Sciences

Language: English

Topic: Smart Agriculture and AI

Subfield: Plant Science

Field: Agricultural and Biological Sciences

Domain: Life Sciences

Open Access status: bronze

Funder: National Key Research and Development Program of China

Award: G2166426431