World record in Maricopa, Arizona

Maricopa, Ariz.–The Lemnatec Field Scanalyzer at the USDA Arid Land Research Station in Maricopa, Ariz., features a 30-ton steel gantry that moves autonomously along two 200-meter steel tracks, continuously imaging the diverse crops growing below cameras and sensors; According to the WORLD RECORD ACADEMY, it sets the world record as the largest robotic field scanner in the world.

“Fixed on a 30-ton steel gantry moving on 200-metre steel rails over 1.5 hectares of energy millet, the high-throughput phenotyping robot continuously captures and maps the growth and development of the crop, generating a massive data stream with an extremely high resolution — about 5 terabytes per day,” the official website reads.

“The Maricopa Agricultural Center looks like a farm, but it’s actually a laboratory. Having the field scanner here is part of our transformation into the next phase of farming,” said Shane Burgess, UA vice president of Agriculture, Life & Veterinary Sciences, and Cooperative Extension; Dean of UA College of Agriculture and Life Sciences; and director of the Arizona Experiment Station.

“The LemnaTec Scanalyzer is the world’s largest crop data collection platform,” said Burgess. “It’s the vanguard of systems that integrate phenotype with genotype to improve agricultural production.”

“The test fields include 176 lines, varieties and hybrids of sorghum planted in an area the size of a soccer field. Approximately 1.25 acres (30,000 to 40,000 plants) will be scanned, with data feeding into the on-site Maricopa Phenomics Center, a joint collaboration with the USDA-ARS Arid-Land Agricultural Research Center and MAC. The University of Illinois handles the big data analysis,” the official website says.

“At 70 feet tall, 92 feet wide and 1,200 feet long, the Field Scanalyzer is the largest agricultural robot in the world, the project’s researchers have claimed. Resembling a gantry with a sensor box positioned in the middle, it tends two acres a day of crops including sorghum, lettuce and wheat while a cluster of cameras and 3D scanners assess their temperature, shape and color, and the angle of each leaf,” reports Robotics & Innovation.



“Field Scanalyzer, funded by the US Department of Energy and the Bill and Melinda Gates Foundation, is currently being tested in Maricopa, Arizona, from where it will send up to 10 TB of data per day to computers in Missouri and Illinois. Scientists at George Washington University and St. Louis University are using AI systems, including machine learning and deep learning algorithms, to analyze and make inferences about the plants.

“The innovation is described as “an automated robotic field phenotyping platform for detailed crop monitoring”. Phenotyping is the process of predicting a crop’s phenotype — the observable physical characteristics of an organism — using only genetic information collected through genotyping or DNA sequencing.”

“The 70-foot-tall behemoth, dubbed the Field Scanalyzer, is the world’s tallest agricultural robot, say the project’s researchers. Resembling an oversized scaffold with a box at its center, it hauls over 2 acres of crops a day, including sorghum, lettuce and wheat, its array of electronic eyes assessing their temperature, shape and color, the angle of each leaf,” reports Wall Street Journal.

“The Scanalyzer beams this data – up to 10 terabytes per day, equivalent to about 2.6 million copies of Tolstoy’s War and Peace – to computers in Illinois and Missouri. Analyzing the range and depth of the generated data is only possible with machines. Learning algorithms, according to data scientists at George Washington University and St. Louis University, where researchers teach the computers to identify connections between specific genes and plant traits that the Scanalyzer observes.

“Deep learning, a form of AI that uses inferences from data to further refine a system, may also help pinpoint how some plant varieties may differ from one another in ways plant scientists may not anticipate, researchers say .”

“The project was originally funded by the US government, and the soccer field-sized robot is 70 feet tall (about 21.3 m), 92 feet wide (28 m), and 1,200 feet (366 m) long, the researchers said in a report on the Rothamsted Research website,” reports Farmer’s Weekly.

“The tall metal structure supports a motorized measurement platform with multiple sensors and is capable of monitoring plants over an area of ​​15 x 120 m. It is also fully automatic and can be operated 24 hours a day, all year round.

“The machine facilitates data collection through sensors that simultaneously and non-destructively analyze plant growth, morphology and physiology, and the data obtained provides a detailed report on plant growth and vigor.”

Field scanner system for phenotyping

“The Lemnatec Field Scanalyzer at the University of Arizona’s Maricopa Agricultural Center and USDA Arid Land Research Station in Maricopa, Arizona, is the world’s largest crop analysis robot that autonomously moves along two 200-meter-long steel rails, continuously accompanying the crops growing below a variety of cameras and sensors,” reports TerraRef.org

Datastores used: weather, sensor and property data

Automated controlled environment phenotyping

The Bellwether Foundation phenotyping facility is an air-conditioned 70m² growth house with a conveyor belt system for transporting plants to and from fluorescence, color and near-infrared imaging cabinets. This automated, high-throughput platform enables repeated non-destructive time-series image acquisition and multi-parametric analysis of 1,140 plants in a single experiment.



Data stores used: sensor and property data

Phenotyping of sensors on unmanned aerial vehicles

UAVs repeatedly fly precise overflights over crops to create autonomous image captures, with onboard sensors geo-registering the data and merging the captured images based on vehicle positions. The deployment of UAVs will allow rapid coverage of areas much larger than what can be covered with ground-based systems.

Data stores used: sensor and property data

Phenotyping of sensors on manned ground vehicles

Multiple vehicles equipped with instruments that measure the canopy’s elevation, temperature and spectral reflectance at three wavelengths are used to collect property-level data that is georeferenced to centimeter accuracy. Plot scale data is relevant for traits that are strongly influenced by interplant competition and plant geometry, such as B. Light interception and crown temperature.

Data stores used: sensor and property data

Genomic and genetic data and computing platform

We will perform whole genome resequencing on a diverse set of 400 sorghum accessions included in the Bioenergy Association Panel to measure the landscape of genetic variation in the selected germplasm and provide whole genome sequences for association mapping. In addition, we will perform genotyping by sequencing on 400 lines from two populations of recombinant inbred lines (RIL). The genomic data is then used to identify the differences between each lineage and the reference sorghum genome sequence. We will use bioinformatics and quantitative genetics to characterize the observed genetic variation and identify genomic regions that control biomass, plant architecture, and photosynthetic traits.

Data stores used: genomic data

High performance data storage and computing system

The system provides public access to raw data, metadata, derived data, provenance of derived data, and standardized data processing workflows. This open-source computing platform is used to provide open access to vast datasets that guide breeding decisions, facilitate collaboration, and enable unprecedented data sharing.

Datastores used: weather, sensor, trait, and genomic data, as well as software (github.com/terraref) and open computing (www.workbench.terraref.org).

Source