Danniell Domingo

Sensors typically used for facial recognition in smartphones have now been adapted to enhance the efficiency of farm robots, making them GPS-independent and capable of navigating varied terrains.


Researchers at the University of California, Riverside, have developed Adaptive Generalized LiDAR Odometry and Mapping, or AG-LOAM, which helps farm robots navigate unpredictable farm terrains.

“The key design principle of this framework is to prioritize the incremental consistency of the map by rejecting motion-distorted points and sparse dynamic objects,” the researchers said.

Unlike previous technologies which heavily rely on GPS signals, AG-LOAM can work accurately in areas where GPS is weak or non-existent.

At its core, this breakthrough technology is powered by adaptive Light Detection and Ranging (LiDAR), where data is produced by the LiDAR sensors as it moves across a terrain.

To significantly minimize inconsistencies in data, AG-LOAM utilizes Generalized Iterative Closest Point (G-ICP) which refines the robot’s position and updates the map accurately.

The G-ICP component, with the help of LiDAR sensors, filters the noisy data caused by obstructions like bumpy land and swaying crops, ensuring that the robot can maintain its accuracy whilst traversing complex agricultural lands.

In strict tests, the researchers used a robot equipped with a LiDAR sensor and real-world data was collected as it moved from different terrain types–flat and rugged lands.

Highly-competent results were found as the AG-LOAM was successful in tracking robot movement and producing precise maps of farms. 

“Results demonstrate that our method can achieve accurate odometry estimation and mapping results consistently and robustly across diverse agricultural settings,” researchers concluded.

However, AG-LOAM may not be affordable for all farmers as it is currently optimized for 360-degree LiDAR sensors.

Given the high production costs and the Philippines' reliance on agricultural imports, fully integrating AG-LOAM into the local agricultural sector could take time.

Millions of Filipinos rely on farming for their food and livelihood, and embracing such technology may require significant investment.

Undeniably, AG-LOAM promises efficient and reliable unmanned farming.