Chonnam National University Researchers Develop New AI System for Highly Accurate Mapping and Modeling of Orchards

Researchers combine aerial maps with ground LiDAR data to create highly accurate digital models of orchards, supporting automation tasks

JEOLLANAM-DO PROVINCE, South Korea, July 22, 2026 /PRNewswire/ -- In precision agriculture, where data is used to automate and optimize agricultural tasks for maximum yield, large-scale mapping and digital modeling of the land of interest are essential, providing the foundation for autonomous navigation and robotic operation. However, in modern commercial orchards, with their large scale, high tree density, and structured layouts, achieving consistent mapping and accurate 3D modeling remains challenging.

While drone-based remote-sensing imagery (RSI), aided by global navigation satellite systems (GNSS), can generate accurate aerial maps, it cannot reliably capture important details beneath dense tree canopies. On the other hand, ground-based robots using LiDAR–inertial odometry can produce high-fidelity 3D point clouds that capture both individual tree structures and orchard row layouts. But weakened satellite signals under dense foliage cause their estimated trajectories to accumulate drift over long distances. Previous studies have explored matching these two data sources to suppress drift and improve accuracy. Conventional matching techniques, however, are highly sensitive to cross-modal differences, seasonal variations, and repetitive row geometry.

To address these limitations, a research team led by Professor Kyeong-Hwan Lee from the Department of Convergence Biosystems Engineering at Chonnam National University in South Korea developed a novel cross-modal fusion framework to integrate low-altitude drone RSI with ground robot LiDAR-inertial measurement unit odometry (LIO). "Our system utilizes deep learning-based cross-modal alignment to connect these two sources of information, identifying the same tree rows, canopy patterns, and open space and creating an accurate model of the orchard," explains Prof. Lee. Their study was made available online on March 26, 2026, and published in Volume 16, Issue 2 of Artificial Intelligence in Agriculture on June 01, 2026.

In the proposed framework, drone RSI collected over an apple orchard was processed into a tiled database, from which an RSI-local map of a specific area was generated. The LiDAR point cloud data obtained from the ground LIO unit was then used to create a structured 2D Bird's Eye View (BEV) local map that preserves essential 3D information. The 2D BEV was then matched with the corresponding RSI local map using a cross-view LIO-RSI fusion matching network. This deep learning model was designed to utilize pixel-level structural cues for reconstructing and accurately aligning the LiDAR map with the aerial map, substantially reducing accumulated drift.

The aligned data was then incorporated into a pose-graph estimation framework to create a geographic information system-driven multilayer orchard model, which can be used to extract a variety of data for orchard management.

In tests, the fusion framework achieved localization accuracy on the order of a few centimeters, demonstrated robustness to seasonal variations, and suppressed long-term drift more effectively than conventional approaches. Moreover, the system is suitable for deployment on embedded devices for real-time operation.

"By integrating what a robot sees on the ground with an aerial map, our system can help agricultural robots work reliably in orchards," concludes Prof. Lee. "In the future, such information could lead to living digital models of farms and orchards, helping farmers adapt to seasons and environmental and societal pressures while producing food more efficiently and sustainably."

Reference
Title of original paper: Transformer-based cross-view LiDAR–orthomosaic fusion for geo-localization and digital modeling in apple orchards
Journal: Artificial Intelligence in Agriculture
DOI: 10.1016/j.aiia.2026.03.005

About the Institute
Chonnam National University (CNU), established in 1952, is one of South Korea's leading national universities located in Gwangju. Building on its founding commitment to cultivating leaders of integrity and professional excellence, CNU contributes to national development and global progress through the pursuit of knowledge, ethical responsibility, and inclusive excellence. Guided by the core motto "Truth, Creativity, and Service," the university advances research, education, and public engagement that strengthen resilient societies, foster sustainable development, and promote the well-being of future generations. As a trusted partner in the global community, CNU remains dedicated to addressing complex challenges in an increasingly interconnected world.
Website: https://global.jnu.ac.kr/jnumain_en.aspx

Media Contact:
Minji Son
82-62-530-5191
[email protected]

SOURCE Chonnam National University