以即時動態定位技術提升作物管理與農機運作效率之應用

張翊庭; 賴俊融; 楊心如; 郭鴻裕; 劉滄棽; Yi-Ting Zhang; Chun-Jung Lai; Hsin-Ju Yang; Horng-Yuh Guo; Tsang-Sen Liu

以即時動態定位技術提升作物管理與農機運作效率之應用

Other Title

The Application of Real-Time Kinematic Positioning Technology to Enhance Crop Management and Agricultural Machinery Efficiency

Journal

精農生機研究與發展學術研討會專刊

Part Of

農業試驗所特刊第245號

Pages

48-55

ISBN

978-626-7651-10-0

Date Issued

2025-01

Author(s)

張翊庭

賴俊融

楊心如

郭鴻裕

劉滄棽

Yi-Ting Zhang

Chun-Jung Lai

Hsin-Ju Yang

Horng-Yuh Guo

Tsang-Sen Liu

URI

https://scholars.tari.gov.tw/handle/123456789/23064

Abstract

本研究探討即時動態定位 (real-time kinematic, RTK) 技術在農業中的應用，旨在提升作物管理的精度以及農機操作效率。RTK 技術是一種基於衛星導航系統的高精度定位方法，能提供公分級的精度，大幅提高農業作業的準確性和自動化程度。透過這項技術，農業經營者可實現包括作物精準施肥、除草、灌溉，以及無人農機導航等多項功能，從而降低人力投入與操作誤差，提升農業生產的整體效能。為驗證RTK 技術的實際應用效果，本研究以即時動態定位技術 (RTK) 為基礎，探討其在精準農業中提升作物管理及農機運作效率的應用潛力。研究涵蓋兩大主題：(1) 利用密度分群演算法 (densitybased spatial clustering of applications with noise, DBSCAN) 處理農機軌跡數據，以準確區分作業農地範圍並有效去除雜訊數據；(2) 結合光達技術進行果樹參數測量與產量估算。研究結果顯示，RTK 定位系統提供高精度的公分級定位，結合DBSCAN 技術能顯著提升農地區域劃分準確性；而光達數據與深度學習模型的應用，可成功估算出香蕉與鳳梨釋迦的產量，並精確分類其採收期與生長期。本研究提供即時動態定位與光達技術的結合在精準農業中的應用範例，為提升農業生產效率及資源管理提供技術基礎。 This study explores the application of real-time kinematic, RTK (RTK) positioning technology in agriculture, aiming to enhance the precision of crop management and the operational efficiency of agricultural machinery. RTK technology, a high-precision positioning method based on satellite navigation systems, provides centimeter-level accuracy, significantly improving the accuracy and automation of agricultural operations. With this technology, agricultural operators can achieve various functions, including precision fertilization, weeding, irrigation, and autonomous machinery navigation, thereby reducing labor input and operational errors while enhancing overall agricultural productivity. To verify the practical applications of RTK technology, this study focuses on its potential to improve crop management and agricultural machinery efficiency in precision agriculture. The research encompasses two main themes: (1) using the density-based spatial clustering of applications with noise (DBSCAN) algorithm to process agricultural machinery trajectory data, accurately delineate farmland boundaries, and effectively eliminate noise; and (2) integrating LiDAR technology to measure fruit tree parameters and estimate yields. The results demonstrate that the RTK positioning system provides high-precision centimeter-level accuracy, and when combined with the DBSCAN algorithm, it significantly enhances the accuracy of farmland boundary delineation. Furthermore, the integration of LiDAR data and deep learning models successfully estimates the yields of bananas and sugar apples while precisely categorizing their harvesting and growth periods. This study offers a practical example of the integration of RTK and LiDAR technologies in precision agriculture, providing a robust technical foundation for improving agricultural productivity and resource management.

Event(s)

精農生機研究與發展學術研討會

Proceedings of A Symposium on Research and Development of Smart Agriculture and Biomechatroincs Engineering

Subjects

即時動態定位

密度分群演算法

精準農業

Real-Time Kinematic positioning

DBSCAN

precision agriculture

Publisher

農業部農業試驗所

Type

conference paper

File(s)

Name

no245-5.pdf

Size

1.24 MB

Format

Adobe PDF

Checksum

(MD5):10ebd3d2525f0ad2f5553934dc8ebe7c