New AI technology combines drones and robots for precise orchard mapping
Advances in precision agriculture are helping farmers improve efficiency and productivity through the use of advanced technologies. One of the key requirements for automated farming is the ability to create accurate maps and digital models of agricultural land. These models support autonomous machines and robotic systems that perform important farm operations.
Orchards present unique challenges for mapping because they often cover large areas, contain dense tree canopies, and have highly organized row structures. Traditional aerial mapping methods using drones can provide detailed views from above but may struggle to capture information hidden beneath tree cover.
At the same time, ground-based robots equipped with LiDAR sensors can collect detailed three-dimensional information, but their navigation accuracy can decrease when satellite signals weaken under dense foliage.
To overcome these challenges, researchers developed an innovative system that combines data from drones and ground robots. The technology uses artificial intelligence and deep learning to connect information collected from both sources. By identifying matching tree rows, canopy structures, and open spaces, the system creates a more complete and accurate digital representation of orchards.
The process begins with drone imagery being converted into detailed aerial maps. Ground robots then gather LiDAR point cloud data, which captures important structural details within the orchard. The AI system aligns and combines these datasets using advanced matching techniques that analyze patterns and structural features. This approach significantly reduces navigation errors and improves map accuracy.
The integrated information is then used to create a multilayer digital orchard model. Farmers and agricultural managers can use this model to collect valuable information, monitor orchard conditions, and support decision-making processes. The system can also help improve the performance of autonomous agricultural equipment.
"By integrating what a robot sees on the ground with an aerial map, our system can help agricultural robots work reliably in orchards," said Professor Kyeong-Hwan Lee from the Department of Convergence Biosystems Engineering at Chonnam National University in South Korea. "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."
Testing showed that the technology achieved highly accurate localization, often within just a few centimeters. It also proved effective across different seasons and reduced long-term positioning errors better than conventional methods. In addition, the system can operate in real time on compact devices.
Their study is published in Volume 16, Issue 2 of Artificial Intelligence in Agriculture.
Photo Credit: Chonnam National University