ASSISTIVE CORN BREEDING TECHNOLOGY AIMED AT FEEDING THE WORLD IN 2050

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: TERMINATED
• Funding Source: HATCH
• Reporting Frequency: Annual
• Accession No.: 1005027
• Project No.: ILLU-741-330
• Project Start Date: Oct 31, 2014
• Project End Date: Sep 30, 2019

Project Director
Grift, T.

Recipient Organization
UNIVERSITY OF ILLINOIS
2001 S. Lincoln Ave.
URBANA,IL 61801

Performing Department
Agri Engineering

Non Technical Summary
Norman Borlaug was the recipient of the Nobel Peace Prize in 1970, and he is often regarded the Father of the Green Revolution which saved the lives of a billion people. His main accomplishments were to bring industrial agriculture to India, Pakistan and Mexico, as well as the development of high yielding dwarf varieties of wheat. Borlaug himself acknowledged that a Second Green Revolution is needed for future generations, but it can, in contrast to the First, no longer expand on fertilizer, energy, water, or arable land. In addition, agricultural production will be limited by more frequent natural disasters due to climate change, as well as facing the end of oil. In light of the latter, agriculture is in fact expected to expand its outputs from food, feed and fiber to include bioenergy. Taking these trends into account, the following developments need to take place. 1) Corn varieties need to be developed that can tolerate very high plant populations. The ERA project carried out by seed company Pioneer indicates that the per-plant yield of a modern variety corn plant does not significantly differ from that of a plant in the 1940s. The difference however, is that due to the architecture of the above ground plant as well as the root system, plants can be grown much closer together, which elevates the yield per acre. 2) The per-plant yield of corn needs to be improved further. 3) Corn varieties need to be developed that have a high resistance to biotic and abiotic stresses. 4) Breeders need to be provided with engineering tools that allow them to conduct more experiments faster. In addition, various other areas in the breeding process where engineering and automation technology can be applied need to be explored. The project as defined here addresses item 4).

Animal Health Component

0%

Research Effort Categories

Basic

30%

Applied

30%

Developmental

40%

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
202	1510	2020	100%

Knowledge Area
202 - Plant Genetic Resources;

Subject Of Investigation
1510 - Corn;

Field Of Science
2020 - Engineering;

Keywords

corn breeding

in-field phenotyping

root phenotyping

controlled pollination

Goals / Objectives
In this Hatch project, we will begin work on a variety of projects aimed at increasing corn yield by aiding the breeding process using an agricultural engineering perspective. Each project increases in complexity, and builds on knowledge gained from previous work. The four projects are named EarBot, FieldBot, RootBot and PollBot.

Project Methods
1. Robotic Manipulation of Ears (EarBot) - The area of ear phenotyping so far has had little attention, arguable owing to the labor involved, and the lack of proper image analysis techniques that can be used to give measurements of the ears. Machinery exists that can harvest full ears (such as for sweet corn) and that can remove the husk. What the machine delivers is a bin that contains de-husked ears in random order. These ears will be fed to the robot on a conveyor belt. The task of the robot is to non-selectively grab ears from the conveyor belt, drill a hole in the base of the ear by presenting the ear base to a rotating drill bit, and then place the ear by placing the hole on a spike that is present in the center of a softbox which provides a highly even lighting scene needed for proper imaging of the ear. The number of images needed is currently determined by human counting of the number of rows in the ear, which can range from 14 to 20. This task will be automated through preview image processing. After the door is closed (using a signal that the robot arm has retreated), the spike is rotated automatically and lateral ear images are taken. When the process is complete, the door opens automatically and the robot removes the ear from the spike. Subsequently, it places the ear on a scale for weighing, and then into a shelling machine which separates the kernels from the ear.2. A Robotic System for In-Field Phenotyping (FieldBot) - Currently, this robot features a single processor, infrared and ultrasonic sensors for crop-based guidance, an electronic compass, remote control, as well as RTK-GPS guidance functionality. FieldBot will also adopt skid steering from the earlier AgTracker design which works well in agricultural fields, and allows spin turns. As mentioned, when the corn plants are small, the robot will be guided by RTK-GPS, but when the plants are taller, the robot will be guided using the stalks themselves as markers. FieldBot will also be fitted with a temperature sensor, a soil moisture sensor, and an electrical conductivity sensor. This will aid in feeding localized environmental condition data to the crop models. These types of sensors have been well documented in the literature and those used in this project are commercially available.3. Robotic Root Phenotyping (RootBot) - In order to expedite the root analysis work, a robotic machine such as RootBot is needed that can perform automated root digging and storing in a corn field. Corn root extraction takes place during flowering time, which in the mid-west occurs in late July-early August. During this period, the soil is hard since rainfall is typically not excessive, which alleviates traction problems. Midwestern fields have ample clay and organic matter content which can make removal of the roots challenging. The plants by the time of flowering can be quite tall (up to 2.5 m) and therefore a gantry style robot is needed. The gantry straddles up to 4 corn rows (3m), and it will be fitted with a vertical robotic arm that can reach the soil level. The guidance of the robot will take place by RTK-GPS since the gantry style offers an unrestricted aerial view of a GPS antenna. Within the gantry coordinates, the target corn stalk location is roughly known from the FieldBot data, but the final stalk location will be fine-tuned using a slanted laser sheet arrangement design from earlier research. Once the plant is localized and identified, RootBot will cut the stalk to approximately 30 cm above the soil, and dig out a 30*30*30 cm cube of soil that contains the root core. After digging, the robot pulls the root from the soil, labels the stalk using a barcode and stores the root cores in a carrousel (estimated up to 100). When the robot has collected and labeled the roots, the root cores are delivered to a cleaning station where they are cleaned. They are subsequently run through an imaging chamber (similar to the Corn Imaging Box used for ear phenotyping) to collect pertinent root data. The imaging chamber can contain cameras from monochrome for morphological feature detection to multi- and hyper-spectral cameras for secondary feature detection.4. Robotic System for Controlling Pollination (PollBot) - We propose to develop a robotic platform that conducts the pollination process automatically. This is a very challenging task, that may not be completed in the term of the current Hatch proposal, however, it is the holy grail of corn breeding technology that will show where the limitations are, but also what potential new technologies can accomplish.

Progress 10/31/14 to 09/30/19

Outputs
Target Audience:College students were targeted in formal classroom settings, high school students were targeted during departmental visitssuch as the Illini Summer Academy, Research Apprentice Program II (minority program), and EYO (Explore Your Options)program as well as ExplorACES (College of Agricultural, Consumer and Environmental Sciences) and College ofEngineering Open House. PI Tony Grift has also organized the ASABE Robotic Student Design Competition from 2007 to2016.Farmers were informed at ACES Agronomy Day, and at invited presentations at meetings.International audiences were reached during various invited presentations from theUnited Kingdom, The Netherlands, and in South Africa.Colleagues from academiawere targeted in various presentations, ranging from workshops to international annualmeetings, journal publications, and ASABE Resource Magazine as follows: Grift, T.E., N. Engeseth, Guest Editors. 2019.Nourish the planet.Special Issue November/DecemberofResourceMagazine, ASABE. Grift, T.E.2018.Thinking outside the furrows: Startups are changing the world., November/December issue ofResourceMagazine, ASABE. Grift, T.E.2018.Climate change: Is it real?The Last Word, July/August issue ofResourceMagazine, ASABE. Grift, T.E.2015.Student robotic competition thriving.Nov/Dec issue ofResourceMagazine, ASABE. Grift, T.E.2015.Pay me, pal! The Last Word, Nov/Dec issue ofResourceMagazine, ASABE. Grift, T.E.Guest Editor. 2015Feed the World in 2050.Special Issue Part 2: March/April 2015 issue ofResourceMagazine, ASABE. Grift, T.E., M.O. Bohn, Guest Editors. 2014Feed the World in 2050.Special Issue Part 1: November/December 2014 issue ofResourceMagazine, ASABE. Changes/Problems:This project was overambitious, and we did not succeed in securing funding to develop the robotic methods proposed. However, we did continue developing assistive methods for corn breeding, with students and faculty involved from Crop Sciences and Agricultural & Biological Engineering. What opportunities for training and professional development has the project provided?This project has provided ample opportunity for training of graduate students, but also farmersand companies. We have demonstrated our root imaging system and worked with companies such and Monsanto,360Yield, and Precision Planting to analyze their roots. We have also branched out our method to USDA-ARS (Marty Williams)that is using it to measure corn plant lodging. How have the results been disseminated to communities of interest?We have demonstrated our methods in various workshops, as well atAgronomy Dayand at the Corn Breeder School. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? The EarBot project was addressed, as we developed a methodology to perform rapid phenotyping of corn ears. The method allowed for automated measurement of up to sixteen morphological parameters of each kernel. In addition, kernel maps were created which could become key tools in ear phenotyping. Grift, T.E., M.A. Momin, Y. Zhang, and M.O. Bohn. 2017.Semi-automated, machine vision based maize kernel counting on the ear.Biosystems Engineering 164:171-180. The RootBot was also addressed in that we refined our Corn Root Imaging Box that was used by various graduate students. Hauck, A.L., J. Novais,T.E. Grift, and M.O. Bohn, 2015.Characterization of maize (Zea mays L.) root system architecture and complexity in a diverse set of ex-PVP inbreds and hybrids.SpringerPlus, 4:424, DOI: 10.1186/s40064-015-1187-0 We have also extended the image analysis software to measure root lodging in field and sweet corn (publication in progress). Grift, T.E., N. Hausman, M. Williams, M. Bohn. 2020. Machine vision based assessment of root lodging in maize. Biosystems Engineering. (In Progress). We are currently conducting our Field / Coffee shop / Computer (FCC)study, where we compare three levels of corn root angle measurement being: 1) human Field assessment, 2) human Coffeeshop assessment based on root images on a tablet, and 3) Computer calculation. We expect thatthe computer is more consistent than humans and that there is little benefitto have the actual root in one's hand; flipping through images on an IPad in a coffee shop will do fine. Once we get the stats worked out, we hope that companies and universities alikewill at least try to automate field phenotyping as much as possible, not just to alleviate graduate students, but also to improve phenotyping consistency and throughput. Grift, T.E., M.A. Momin, R. Reis, R. Shaheb, A. Sarker, I. Ono, and M. Bohn. 2020. Computer versus human assessment of maize root top angle. Biosystems Engineering. (In Progress). We are currently in discussion with the National Institute of Standards and Technology (NIST)to start an effort of standardization within crop sciences. The main argument from NIST is that USDA has never contacted NIST to help them in their standardization efforts. We are writing an opinion paper for the journal Phenotyping where we propose to develop astandard development and maintenance infrastructure akin to our ASABE Standards Committee structure. Our 2011paper will be used as an example of how algorithms could be used as standards. Grift, T.E., M., Bohn, and J. Novais. 2011.High-throughput phenotyping technology for maize roots.Biosystems Engineering 110, 40-48, DOI:10.1016/j.biosystemseng.2011.06.004. We were unable to secure funding for the FieldBot and PollBot concepts.

Publications

• Type: Journal Articles Status: Awaiting Publication Year Published: 2020 Citation: Grift, T.E., N. Hausman, M. Williams and M. Bohn. 2020. Machine vision based assessment of root lodging in maize. Biosystems Engineering. (In Progress).

Progress 10/01/17 to 09/30/18

Outputs
Target Audience:The target audiences ranged from researchers, in both agricultural robotics and crop sciences (high-throughput phenotyping), to students and thegeneral public. Changes/Problems:The High-Throughput-Phenotyping research will continue because we do not need extra funding, we have Corn Root Imaging Boxes (CRIBs), it is a matter of upgrading software and develop new imaging processing algorithms. There is a major turn in the project direction, because I have been able to gain funding for soil sensing research (EBI, Michelin, and Kelly Engineering). Therefore the big push will be in the direction ofOTIISS/SNIFF. The robotics research has come to a halt. Our department hashired a new faculty member who is doing very well;I have decided to let him take the lead on robotics research, and let me pursue other opportunities. What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest?We have communicated the FCC project and OTIISS concept at Agronomy Day inUrbana, Illinois. What do you plan to do during the next reporting period to accomplish the goals?We will first publish the results from the FCC and root lodging projects, we have all the data already. We will continue the development of the OTIISS/SNIFF idea, and apply for a large grant from USDA/AFRI. We will also publish several papers from our Michelin funded tire impact on soil and crop project (one PhD student is finishing by June 2019). We will complete a project funded by the Energy Biosciences Institute which allowed us to initiate the OTIISS concept. We are in the process of applyingfor a provisional patent on the OTIISS concept.

Impacts
What was accomplished under these goals? Using Hatch funding, we have begun investigation on the quantification of corn root lodging in collaboration with USDA employees Dr. Marty WIlliams and Nick Hausman (both at Illinois). I have written software that builds on our previous methods (seeGrift, T.E., M., Bohn, and J. Novais. 2011.High-throughput phenotyping technology for maize roots.Biosystems Engineering 110, 40-48) but now also measures the lodging angle of sweet corn compared to field corn. We have also completed data collection for our Field/Coffeeshop/Computer (FCC) project, where we measure the corn root angle (we omit the root complexity which we express in the fractal dimension, that is near impossible for humans to replicate). We compare Field measurement (human observing a root while holding it in her/his hand), to Coffeeshop (human observing images of corn roots on a tablet), and acomputer algorithm (according to our 2011 definition). We hypothesize that the computer will be most precise (consistent) but not necessarily accurate, whereas we assume that humans introduce error due to subjective reasoning and the learning effect (observations biased by previous ones). We have contacted a cognitive psychologist to help us in the data analysis process and publication. In addition, we have started a new project titledOn-The-Implement-Intelligent-Soil Sensing. We are working on the development of thenew generation of retrofittable, on-the-go soil sensing technologies that go far beyond what is possible today. We already have an ongoing research program in which we are exploring mechanical force, active and passive acoustics, electrical capacitance, electrical impedance and shear wave technology (independent of moisture, unique in soil sensing). We are also pursuing the Holy Grail in soil sensing, being the measurement of plant available nitrogen based on spectroscopy ofsoil fumes (the title of the upcoming proposal is Sensing Nitrogen In Farm Fields or SNIFF). Apart from soil data collection at any time a farmer enters a field, these developments will allow soil scientists to properly calibrate their models. We have submitted a proposal for seed funding, and we will target USDA-AFRI to fund the complete project. We have also graduated a MS student Brandon Tate who developed a complete Vertical Acoustic Cone Penetrometer method.

Publications

• Type: Journal Articles Status: Published Year Published: 2018 Citation: Gan, H., S. Mathanker, M.A. Momin, B. Kuhns, N. Stoffel, A.C. Hansen and T.E. Grift. 2018. Effects of three cutting blade designs on energy consumption during mowing conditioning of Miscanthus Giganteus. Biomass and Bioenergy 109:166-171.
• Type: Journal Articles Status: Accepted Year Published: 2018 Citation: M.A. Momin, T.E. Grift, Valente, S. and A.C. Hansen. 2018. Sugarcane yield mapping based on vehicle tracking. Precision Agriculture.

Progress 10/01/16 to 09/30/17

Outputs
Target Audience:The target audience ranges from researchersin both agricultural robotics and crop sciences (high-throughput phenotyping) tostudents and the general public. Changes/Problems:The main problem is the lack of funding for robotics, especially applied in interdisciplinary projects where there always seems to be confusion about the aim of the project. Everyone claims that interdisciplinary research is highly valuable, but our experience is that panels never seem to understand the power of the synergy between engineers and crop scientists. Our current tenure process also discourages inter-disciplinary work since it is hard to evaluate the contribution of individual researchers. Nevertheless, we will continue our quest to convince crop scientists to stop avoiding the "dark side" of the plantbecause it is laborious and cumbersome. The root system of any plant is the only organ in contact with soil, water, air and nutrients, and it needs to be studied in equal proportion to the above ground plant. More funding is needed to pursue this aim. What opportunities for training and professional development has the project provided?The results from our most recent root work will be presented at the CSS & Seed Expo, Chicago, Illinois in December of2017. How have the results been disseminated to communities of interest?Three press releases have been submitted: one about our paper on Mango quality, one on soybean quality, and one on the corn ear imaging research: https://news.aces.illinois.edu/news/new-machine-evaluates-soybean-harvest-quality https://aces.illinois.edu/news/new-method-analyzes-corn-kernel-characteristics https://dailyillini.com/news/2017/11/19/researchers-develop-breakthrough-analyzing-corn-kernels-illinois/ What do you plan to do during the next reporting period to accomplish the goals?We will continue to refine the corn ear analysis method; depending on funding we would like to evaluate whether nutritional composition of individual kernels and its distribution on ear is feasible. Also, we will continue work on improving and extending the capability of our corn root analysis methods to lodging. We are planning to design a completely new experiment to evaluate whether the current technology will be able to predict which genotypes are most susceptible to lodging.

Impacts
What was accomplished under these goals? We have finalized the development of the FieldBot;it is currently being used to determine the quality of turfgrassin terms of playability and safety. This robot goes beyond the typical machine vision based phenotyping robot, in that it not only uses imagery to determine the texture of turfgrass, but it also measures mechanical friction in a linear and rotary fashion. We hope that this robot will not only have an impact on agriculture, but also in sport fields, where the current evaluation of playability and safety is primitive to say the least. Allthough we have not siginificantly improved the handling of corn ears, we have refined the algorithms that allow for determination of morphological parameters of individual corn kernels, and we have published a paper showing root maps (scrolls) that the experts claim have never been seen before. We have also furthered the corn root quality assessment method significantly. In 2017 we have collected 1,500 corn roots that we evaluated in three ways. Firstly we have collected traditional images, then three human evaluators estimated the Root Top Angle (RTA), and thirdly, the same evaluators estimated the RTA using the imagery on a computer screen, conceivably in a coffeeshop. We call this the FCC project which stands for Field/Coffeeshop/Computer method. The aim is to show universities and companies alike that there are major advantages in imaging plant phenotypes and evaluate traits either using a computer algorithm (if available) or by human observers. We are working together with Dr. Aaron Benjamin who is a professor in Cognitive Psychology. We have also made significant progress in attempting to quantify lodging from root images. We are working with USDA's Dr. Marty Williams and we have defined various new methods to quantify lodging in roots and we will cross correlatethe data with lodging scores from the field. We anticipate several papers from this effort, where the first will be engineering methods oriented (similar to our 2011 paper titled "High-throughput phenotyping technology for maize roots") followed by a paper where we compare sweet corn and field corn in terms of lodging potentialand possibly other traits.

Publications

• Type: Journal Articles Status: Other Year Published: 2017 Citation: Momin, M.A., M.T. Rahman, M.S. Sultana, C. Igathinathane, A.T.M. Ziauddin and T.E. Grift 2017. Geometry-based mass grading of mango fruits using image processing. Information Processing in Agriculture 4:150-160. DOI: 10.1016/j.inpa.2017.03.003.
• Type: Journal Articles Status: Published Year Published: 2017 Citation: Momin, M.A., K. Yamamoto, M. Miyamoto, N. Kondo and T.E. Grift. 2017. Machine vision based soybean quality evaluation. Computers and Electronics in Agriculture 140, 452-460.
• Type: Journal Articles Status: Under Review Year Published: 2017 Citation: Hauck, A., J.H. Kim, M.H. Kim, B. Wang, W. Song, M.O. Bohn, T.E. Grift and J. Lai. 2018. Unearthing root architecture QTL in a population derived from the elite Chinese maize hybrid ZD958 G3 (Submitted).
• Type: Journal Articles Status: Published Year Published: 2017 Citation: Grift, T.E., W. Zhao, M.A. Momin, Y. Zhang and M.O. Bohn. 2017. Semi-automated, machine vision based maize kernel counting on the ear. Biosystems Engineering 164:171-180, DOI:10.1016/j.biosystemseng.2017.10.010.

Progress 10/01/15 to 09/30/16

Outputs
Target Audience:The target audience includesresearchers, in both agricultural robotics and crop sciences (high-throughput phenotyping), as well as students and the general public. Changes/Problems:The need for high-throughput phenotyping is more relevant than ever; however, the biggest challenge has been finding federal funding. We have proposed many ideas in the past (for instance a robot for corn nurseries) to the NRI program which were considered too ambitious. Nevertheless, such an exact robot development was funded in 2015. It seems that much of the research we proposed a decade ago, whether in agricultural robotics or high-throughput phenotyping, has a lag time for 10 years before funding becomes available. However, our philosophy is that we need to work on big problems to Feed the World in 2050. Although it is tempting to go after funding that keeps the academic system going, just incremental small steps tend to ingore the ultimate target over time. Although considered popular publications, the special issues produced by the PI and Dr. Martin Bohn titled "Feed the World in 2050" are a testimony to our caring for future generations rather than our current status quo. What opportunities for training and professional development has the project provided?The funding has allowed us to disseminate the findings of the high-throughput phenotyping to professional groups such as the ASABE chapter of the Quad Cities in Illinois. How have the results been disseminated to communities of interest?We have used the funding to dissminate findings at farm shows as well as the Agronomy Day or the University of Illinois. What do you plan to do during the next reporting period to accomplish the goals?We will continue the work with Tryston Beyer, Dr. Randy Nelson, and Dr. Andrew Hauck, who are currently running a breeding program in China. We are also looking forward to working with the two PhD students in Mathematics, as they have a unique perspective on the matter of root morphology. The Michelin funded project will continue for two more years and we will make ample use of our CRIBs to assess further the influences of compaction on root development. The 3D imaging system has been on the backburner for a while, but if students are interested, we will revive this project as well. Although we have a working small size field robot, we are lacking the funding to develop new sensors for it, but we are pursuing a project with Co-PI Dr. Martin Bohn through the NSF/USDA NRI program along these lines.

Impacts
What was accomplished under these goals? In the area of the EarBot, we have developed a new imaging system that allows us to obtain images of corn ears, both a complete ear as well as a detail in the center of the ear. We have developed software that separates each kernel in the ear, which has been quite a difficult task even under controlled diffuse lighting in our dedicated "soft box". Once the images of individual kernels are available, we calculate up to 16 morphological parameters of each kernel (the most prominent being location in the ear, perimeter, area, and equivalent diameter). These parameters are currently being evaluated for their use in Quantitative Loci Mapping by Crop Sciences collaborator Dr. Martin Bohn and his group. The aim is to find out which genes in the maize genome are responsible for the trait being measured. The imaging system also allows for creating ear maps which clearly show the kernels in the ear, but until now we have not gone beyond creating beautiful images. We have also created a new Corn Root Imaging Box (CRIB) to be used as a 3D root analyzer. Here a corn root is mounted upside down, and a slanted laser beam is passed by it. The areas lit by the laser sheet are being detected by a camera, and since we know the exact laser sheet angle, we can determine where this lit area is located in a 3D space. This will allow us to create rudimentary 3D models of corn roots in an attempt to further generate more data for QTL analysis. In the area of the FieldBot, we have created a new robot to be used as a ground-based platform for field measurements in 2017 and beyond. The new robot is much lighter than earlier versions, and it has been fitted with the latest in electronics for sensing and guidance. We are now implementing a highly precise but low-cost RTK-GPS system for its guidance. We have also experimented with drones for phenotyping in experimental plots, but we have come to the conclusion that drones are too limited to be used in experimental plots, they cannot penetrate a canopy, and even the slightest wind will make precise measurements impossible. We leave that kind of research to those who wish to work on a field level, rather than a plot level. We have also worked with two PhD students in Mathematics, guided by Dr. David LeBauer, to start producing mathematical models that accurately describe the morphology of corn roots underground. In addition, we have worked with graduate student Tryston Beyer, whom we have lent a CRIB to study the influence of stripped nitrogen applications on the development of roots. We are using the same method to determine how roots develop under compaction (as part of a Michelin funded project where we evaluate the influence of their low ground pressure tires). We are also working with Dr Randall Nelson (USDA) to extend the root work to soybean, which has proven to be quite difficult in the past.

Publications

• Type: Journal Articles Status: Published Year Published: 2015 Citation: Hauck, A.L., J. Novais, T.E. Grift and M.O. Bohn, 2015. Characterization of maize (Zea mays L.) root system architecture and complexity in a diverse set of ex-PVP inbreds and hybrids. SpringerPlus, 4:424, DOI: 10.1186/s40064-015-1187-0.
• Type: Journal Articles Status: Submitted Year Published: 2016 Citation: Hauck, A., J.H. Kim, M.H. Kim, B. Wang, W. Song, M.O. Bohn, T.E. Grift and J. Lai. 2016. Unearthing root architecture QTL in a population derived from the elite Chinese maize hybrid ZD958. G3 (Submitted).

Progress 10/31/14 to 09/30/15

Outputs
Target Audience:Members of the target audience included the scientific community, farmers, and students. Changes/Problems:The major problem is that limited funding is available for the kind of research we propose. Our proposals are typically multi-disciplinary and often it seems unclear to review panels what the project is about. Although we all claim that interdisciplinary research is important, the funding mechanism (or academic promotion system) does not encourage it. What opportunities for training and professional development has the project provided?We have demonstrated the root characterization technology at various venues such as the corn breeder school, as well as our College of ACES Agronomy Day. How have the results been disseminated to communities of interest?Apart from the scientific publication, we have written popular articles such as the cover article in Resource Magazine titled Maize Root Matters. Together with crop scientist Martin Bohn we have produced two special issues of the ASABE Resource Magazine titled Feed the World in 2050 issue 1 and issue 2. The issues contain 50 articles written by visionaries such as two Food Prize winners, as well as Deputy Secretary of Agriculture Krista Harden. What do you plan to do during the next reporting period to accomplish the goals?We currently are developing a field robot to accomplish the FieldBot goal. We are also writing proposals to the National Robotics Initiative to start work on the PollBot (pollination robot).

Impacts
What was accomplished under these goals? We have attempted to automate the process of phenotypingcorn ears as a starting point for EarBot. One student has graduated with an MS degree on this topic. We have also expanded research in corn roots where we have investigated the root complexity and root angle among a large range of hybrids as a starting point for RootBot.

Publications

• Type: Journal Articles Status: Published Year Published: 2015 Citation: Hauck, A.L., J. Novais, T.E. Grift and M.O. Bohn. 2015. Characterization of Maize (Zea mays L.) Root System Architecture and Complexity in a Diverse Set of Ex-PVP Inbreds and Hybrids. SpringerPlus, 4:424, DOI: 10.1186/s40064-015-1187-0.
• Type: Theses/Dissertations Status: Published Year Published: 2014 Citation: Zhao, W. 2014. High-Throughput Phenotyping Technology for Corn Ears. MS Thesis - University of Illinois. http://hdl.handle.net/2142/72828.