Date: January 2024
Authors: Paige Beuhler and Ines Hanrahan
In December of 2023, the Washington Tree Fruit Research Commission (WTFRC) approved $328,077 for four (4) new technology research projects for the upcoming year. All newly approved technology projects this year are funded 100% by apple.
2024 New Technology Project Details
Project Title: Granny – Image Based Analysis of Fruit Quality
Organizations: USDA-ARS, Washington State University (WSU)
Principle Investigators: Honaas, L., Ficklin, S.
Total Funding Amount for All Years: $151,120
Length: Two (2) Years
This project is a continuation of funds to further develop a software package named Granny that has already been started. The prototype software uses artificial intelligence, machine learning (AI/ML) and image processing to identify individual fruit on a tray and then rate the fruit for peel defects, various peel color attributes, and starch content of fruit disks (using industry standard starch index cards). Currently, Granny will rate stained fruit disks for starch content, whole fruit for superficial scald, peel color, and blush, and provide high-resolution ratings. When industry accepted indices are used, Granny will provide a high-resolution rating, plus assignment to the corresponding category or bin on the current index. The continuation of the project aims to give growers the tools to assess and make critical decisions. This software will reduce rater bias and ensure reproducible assessments of fruit. Higher resolution ratings based on mathematical models should provide higher accuracy. Removal of human-rater bias will allow trait ratings to be comparable across orchards, years, locations, and climates. Comparable, high resolution data collected long term is what is needed in the Smart Orchard to support development of improved decision support systems that use fruit trait data coupled with climate data. High quality data means more opportunity for new tools. The team plans to convert Granny to a model-viewer-controller design, allowing for a smoother end user experience.
Project Title: 3D Orchard Mapping and Renewal Pruning Using a Drone and Ground Robot
Organizations: The University of Tokyo, Oregon State University
Principle Investigators: Guo, W., Davidson, J., Blok, P., Burridge, J.
Total Funding Amount for All Years: $300,000
Length: Three (3) Years
The goal of this project is to develop a prototype robotic pruner that will accomplish dormant season renewal pruning. Dormant season renewal pruning is a sensible first step for automation since it targets relatively simple cuts and thicker wood. The smaller, more delicate cuts, requiring more complex decision-making and a higher level of dexterity, will be delegated to a leaner workforce. The team will target dwarf sweet cherries grown in upright fruiting offset (UFO) systems and secondarily dwarf apples in vertical or tatura trellis systems. There is potential to expand to other trellising systems producing a fruiting wall. The data and knowledge generated, as well as the experience, will serve as a foundation for subsequent efforts to automate other pruning tasks, including the finer cuts.
The overall objective is to develop and test an autonomous robot that can perform dormant season renewal pruning. This has been divided into three sub-objectives:
- Collect data from fruit tree orchards in Washington State and optimize data processing pipeline.
- Develop algorithms to determine renewal cut locations for the robotic pruner.
- Develop, deploy, and evaluate a first robotic pruner prototype in Washington apple and cherry orchards.
Project Title: Smart Apple Orchard: Technology Testbed and Demonstration Site
Organizations: Washington State University (WSU), Innov8Ag
Principle Investigators: Khot, L., Sallato, B., Mantle, S., Peters, T., Kalyanaraman, A.
Total Funding Amount for All Years: $195,901
Length: Three (3) Years
The WTFRC funded Smart Orchard initiative (to Innov8Ag and Washington State University [WSU]) has been a great success in bringing together public-private technology developers to address grower needed solutions. Setting the Smart Orchard in a commercial orchard has been key to provide a realistic testbed for the sensors/technologies evaluation and enhance the extension and outreach to agricultural community. Stemming from these past efforts, the team will now focus on a “Smart Apple Orchard: Technology Testbed and Demonstration Site” to be established at a centrally located cooperating grower block near Mattawa, Washintgon, to continue these efforts tied directly to relevant WSU-lead efforts by the team through the AgAID Institute, AgWeatherNet, Center for Precision and Automated Agricultural Systems, and WSU Tree Fruit Extension.
The team will continue to push for technology evaluation, through this and affiliate projects in the testbed, including 1) environmental indicators monitoring and cold/heat stress mitigation, 2) soil and plant nutrient mapping and management, 3) irrigation automation, 4) precision chemical applications, 5) crop mapping and blossom/yield estimation, and 6) robotic pruning/thinning, weed management, etc. in the next three years. WSU principle investigators with respective expertise will collect relevant ground-truth data to validate and report on the technology. This information is considered critical for grower understanding of technology applicability, as well as for the research team to develop data-driven models for on-farm decision support.
Project Title: Robotics Ready Data Standard for Washington Apples
Organizations: The Yield, Washington Tree Fruit Research Commission (WTFRC), Robotics Plus Limited, Yamaha
Principle Investigators: Harvey, R., Hanrahan, I., Saunders, S., Paul, N.
Total Funding Amount for All Years: $75,000
Length: One (1) Year
The apple industry uses various ag-tech software systems, captured in different vendors’ systems, but currently lacks unified data guidelines to allow interoperability which is essential for data analytics and artificial intelligence/machine learning (AI/ML). During this project, The Yield intends to accelerate the alignment of data guidelines by working collaboratively with WTFRC, multiple Yamaha portfolio companies, and the apple industry’s key technology companies. The goal is to align on data standards and ensure that the industry is collecting data in a way that enables analytic insights, field robotics and predictive modeling. By collaborating with industry and with a commitment to an open system, the partners’ goal is to increase the speed at which the industry realizes value from technology adoption and prepares for the future use of robotics with a focus on Multi-Tool Autonomous Vehicles (MTAV) and crop load management.
The first objective is to conduct a survey of Washington apple industry technology software solutions with a focus on data required to create analytics solutions to improve crop load management and thinning outcomes.
The team will work collaboratively with WTFRC, soliciting input from knowledgeable experts, to design an online survey instrument for apple producers that addresses the following questions:
- Which software technology vendors do you utilize?
- Over how many acres are these tools applied?
- What kinds of data are input into or generated by these solutions?
Team members will communicate extensively with industry via newsletters, extension events and other channels to convey the purpose of the survey. This will be essential to generating the strong producer response needed for an accurate assessment of the most important software technology vendors.
Contact:
Paige Beuhler, Administrative Officer
paigeb@treefruitresearch.com
509-665-8271 ext. 2
Ines Hanrahan, Executive Director
hanrahan@treefruitresearch.com
509-669-0267
