Home / Cases / 12 Potential Uses for UAS in Agriculture
uas

12 Potential Uses for UAS in Agriculture

In North Dakota, there are 82 billion reasons the state is interested in unmanned aircraft systems (UAS). Over the next 10 years, many of those reasons could translate into real revenue. Precision agriculture, experts say, stands to capture the lion’s share of that market.

Although this technology would seem to hold great promise for farmers and agribusinesses alike, there is little evidence to support that its flight path is on course.

“One of the major stumbling blocks in the adoption of UAS in agriculture is the lack of proof of concept and methodologies for adopting UAS data in crop and livestock management,” says John Nowatzki, agricultural machine systems specialist, North Dakota State University. With the launch of a project at the University’s Carrington Research Extension Center (CREC) in 2014, researchers evaluated the usefulness and effectiveness of UAS in crop and livestock management.

“Currently, there are many producers, consulting companies, and small precision agriculture industries in North Dakota interested in adopting UAS technology,” he says. “Our research will provide the information and tools necessary for these companies to successfully use UAS technology in agriculture. Research will also further the technical know-how for applying UAS in agriculture.”

The Center also expects to develop decision-support systems that can be adopted and further refined by UAS industries.

Following are 12 objectives and preliminary findings.

1 Identify plant emergence and plant populations in corn, soybeans, and sunflower. Researchers measured plant emergence and population twice (one week after planting and 12 days after planting) at multiple locations in the field. Multispectral image data collected from UAS during the same time period will be corrected, calibrated, and processed to detect crop plants from the surroundings. Plant stand will be computed for every square-meter area to generate a map of plant stand.

Findings. A Sony NEX 5 camera on the Draganflyer X4-C, with 0.9 cm spatial resolution of multispectral imagery, gathered stand counts on corn. Matlab, a software program, was actually used to do the counting.

“Normalized difference vegetative index (NDVI) collected from a UAS was successfully used to predict the date of soybean physiological maturity within a soybean variety trial,” says Mike Ostlie, CREC. “Corn plant population was measured with a UAS and counted on the ground. We found a good correlation when the corn plants were very small.”

2 Identify any nitrogen deficiencies in corn and wheat. Researchers used optical sensors to establish the NDVI when the corn was in the V-5 and V-8 growth stages at multiple locations in the field. Infrared image data gathered by UAS on the same days will be corrected, calibrated, and processed to detect nitrogen deficiencies.

Findings. The NDVI values generated from the images were compared with the ground truth data to establish the accuracy and ability of the UAS sensing system to identify NDVI values that can be used to apply in-season variable-rate nitrogen. Researchers also collected wheat NDVI data at flowering stage and compared it with the UAS data.

“We compared GreenSeeker data with the UAS-collected data in wheat,” says Ostlie. “Nitrogen deficiency was detected with similar precision in both platforms. The values differed between the two data collection systems, but the differences between the treatments were similar.”

3 Assess early plant health. Researchers will utilize the broad diversity of spring wheat, soybean, and corn studies at the CREC to identify plots with a range of early-season yield potential, nutritional levels, plant vigor, and stand.

Findings. Multispectral images were obtained during the same timeline in which researchers collected ground-based data related to the same parameters measured on the ground.

“Both Greenseeker sensors and aerial imagery could predict relative spring wheat and durum yield potential while the crop was still green,” says Ostlie.

4 Know disease symptoms. Part of a systems approach toward managing disease is having knowledge of disease incidence and severity within a field so producers can better manage that crop or field in the seasons that follow.

CREC collaborators will identify a range of disease severities within research trials. Initial crops and diseases to be investigated will be sclerotinia (white mold) in both soybean and dry edible beans. The general leaf spot complex of tan spot, septoria, and rust in spring wheat will also be investigated and quantified.

Findings. Images were collected from cereal and legume disease trials. Data has not yet been analyzed.

5 Look for insect damage symptoms. Crop producers need to monitor crops for insect damage to determine the needs, locations, and amounts of in-season insecticide applications.

Findings. Insect damage and identification research is still under investigation.

6 Monitor weed infestations. Crop producers need to monitor crops for weed infestations to determine the needs, locations, and amounts of in-season herbicide applications.

Findings. “A handheld GreenSeeker did not provide much information about weed control, and neither did an initial scan of the image taken from the UAS,” notes Ostlie. “However, with the UAS image, a specific area can be targeted. In this case, we looked at weed control in-between the wheat rows and found a nice correlation with the visual estimate of weed control. Weeds that grow in patches and require special control measures, such as Canada thistle and common milkweed, can be mapped for use during herbicide applications or tillage operations.”

7 Notice moisture stress on irrigated crops. Producers who irrigate their crops need to monitor moisture stress to manage irrigation scheduling.

Findings. New thermal imaging equipment will allow water stress monitoring in 2015.

8 Note the impacts of tillage and crop rotations. Producers need to monitor the impacts of tillage and crop rotations on crop emergence, vigor, and yield for use in annual crop selection on their fields.

Findings. Research is ongoing. Results are not yet available.

9 Determine the breeding activity for herd sires and beef females. Breeding success is critical to profitable beef production. Real-time monitoring of breeding activity and movement for bulls and cows make producers aware of mobility problems as well as normal breeding behavior.

Findings. Research is ongoing. Results are not yet available.

10 Take the temperature of animals and the feedlot surface temperatures of various beddings. Feedlot managers need to monitor heat load on feedlot surface temperatures and relate it to actual animal body temps to mitigate stress in animals. Compost piles may be remotely monitored more easily to determine the need for aeration/turning or spreading or marketing finished compost.

Findings. Researchers will measure the surface temperature of the various bedding materials using ground thermometers. Thermal-image data will be collected with a UAS to detect the surface temp of the feedlot. Research is ongoing. Results are not yet available.

11 Detect diseased beef animals in pastures. Beef producers need to identify diseased animals as soon as possible to treat the animals, or isolate sick animals.

Findings. Research is ongoing. Results are not yet available.

12 Identify animals with extreme dispositions. Animals will be monitored in pastures three times each day to identify disruptive behavior. Data from the multispectral images will be compared with the ground truth data to establish the accuracy and ability of the UAS to identify rogue animals. This data collection system will be repeated three times during the summer animal-grazing period.

Findings. Research is on-going. Results are not yet available.

“This industry is still in its infancy,” says Ostlie. “There are bound to be great improvements in the speed of processing data and ease in which it is collected and viewed. While we still have a lot of data to evaluate from this past growing season, we will continue our research efforts in 2015 to see what other information can be gleaned with this new technology.”

(Source – http://www.agriculture.com/technology/robotics/uas/12-potential-uses-f-uas-in-agriculture_587-ar51680)

12 Potential Uses for UAS in Agriculture обновлено: December 21, 2015 автором: admin

Leave a Reply

Your email address will not be published. Required fields are marked *

*