Bringing robots into the field: ABE is developing a robot to analyze crops at a pace equivalent to a tireless crew of 40

Bringing robots into the field: ABE is developing a robot to analyze crops at a pace equivalent to a tireless crew of 40

Plant holder analyzer in large-scale operation

After more than a decade in business, Iowa State University Department of Agricultural and Biosystems Engineering Professor Li Tang has tuned and developed a machine to automate phenotypic measurements of crops in the fields.

PSA was implemented on Syngenta's self-guided UTV platform (Photo provided by Syngenta)
PSA was implemented on Syngenta’s self-guided UTV platform (Photo provided by Syngenta)

With the ability to count eight rows of corn crops at a time, this machine measures crops like corn at an average of 40 tireless people — and with a level of accuracy higher than the average person can retrieve. Known as the Plant Stand Analyzer (PSA), this machine has a set of laser sensors on each crop plant in the eight rows, and if the laser sensors register a crop plant, an algorithm on the computer can detect it — and the scientists can count the number of crop plants, also known as the number of crop plants. The rack, queues in real time, at speeds up to 10 miles per hour. Tang invented this machine in Iowa, and he licenses the technology from the Iowa State University Research Foundation. After licensing, Tang started a small company called “FieldRobo LLC,” dedicated to producing devices like the Plant Stand Analyzer. PSA won the 2019 AE50 Award representing the best innovations in engineering and technology for agricultural, food and biological systems.

This machine alters phenotypic measurements known today – saving time and money in the field.

“In the past, if you want to get the data of plants, such as the number of stands, stem diameter, plant height, leaf angle, leaf area, etc., you would have to go to the field and measure it manually,” Tang said. This has truly become a bottleneck for plant breeding research and plant phenomenology. Now, anyone can drive a machine like a PSA in the field, and then you can record the data from the laser. Every millimeter is measured, creating a profile for each crop row.”

Since the PSA calculates whether or not a crop plant is present, it may seem that it would be easy to accidentally record weeds or other variables as a crop plant. But the machine is so complex that most of the time it can determine if it is an ordinary corn plant, weed plant, or corn plant with tiller, also known as plants with extra buds, or plants with double seeds. There are currently 56 high-speed laser sensors on the analyzer, seven of which collect data from each row. Once the laser beams hit a crop plant in front of the sensor bank, the lights are reflected off the crop plant and bounced back to the sensors, and this reflection is captured 3 times for every millimeter transmitted by the PSA.

PSA in full-scale operation (courtesy of Syngenta)
PSA in full-scale operation (courtesy of Syngenta)

This can be very useful for seed companies that calculate seed germination rates when they are planted, but do not have a large number of technology to see how the seeds develop.

“When seed companies plant research grounds, they know how many seeds they are putting in the ground,” Tang said. But depending on genetics and other factors, not all seeds will germinate. They want to monitor the quality of the data that is being generated. Also, if they don’t know how many plants contribute to the crop, then their final data analysis is skewed.”

But now, this device can show how many plants are growing in each row. As the analyst collects data, the scientists see that each crop plant is represented as a green circle on a computer screen, arranged by row, which will also be organized into charts automatically by custom mapper software. The plant stand analyzer produces not only the number of stands, but also a host of other useful metrics such as stem size, spacing between members, gaps, multipliers, and tillage – which no other high-throughput system can provide even today.

“While the plant base analyzer is very powerful and successful, there are some competing technologies in this field. But when looking at large-scale performance, the main competition is from drone-based systems. However, PSA offers some distinct advantages in Its fast data-to-information turnaround time (same day), superior rack detection accuracy (97% on average), and a large operational time window (3-4 weeks), Tang said, “no special license required to operate the system.”

In the future, this sensing technology could be used in other ways, from analyzers to the many different devices used to collect crop data and give treatments to individual crop plants.

“With the sensor technology used in PSA, we can detect individual plants on the go. Imagine you have this sensor paired with something like a fertilizer spreader, you can regulate the amount of spray and fertilize only when there is a plant in existence and save the chemicals.”

To address the many challenges in phenotyping field plants, and as one of the faculty members at ISU’s Institute of Plant Sciences, Tang has been working closely and proactively with botanists and the plant seed industry to develop robotic solutions, such as his ongoing efforts to develop a PhenoBot that will be able to acquire a range of Advanced features through 3D vision, machine learning and automated manipulation.

#Bringing #robots #field #ABE #developing #robot #analyze #crops #pace #equivalent #tireless #crew

Leave a Comment

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