Date: January 2022
Authors: Paige Beuhler and Ines Hanrahan
In December of 2021, the WA Tree Fruit Research Commission approved $706,830 in new technology research for the upcoming year. All newly approved technology projects this year are funded 100% by apple.
2022 new technology project details:
Project Title: Apple Harvest End Effector and Apple Transport System
Organization (s): Milano Technical Group Inc.
PI (s): Milano, D.; Datta, D.
Total Funding Amount for All Years: $245,000
Length: 2 years
MTG has focused on agricultural operations since 2014. MTG believes that mechanized harvesting, utilizing robotic picking, is one method towards addressing the need to offset the labor shortage while lowering the cost of the overall harvest. The primary objectives of this effort will be:
- Design and build a Robotic Apple Harvester System to include: an Apple Harvest End Effector, arm structure, and an Apple Harvest Transportation (to bin) Subsystem.
a. System ROI goal will be 90-120 days (cost of manufacturing will be approx. 50% of commercially available manual platform costs are)
b. Pick will preserve fruit, tree, and bud integrity
c. Transporter will prevent bruising and puncturing
d. Harvesting subsystem will fit within commercially available platform geometry
e. Arm control system architecture will support common location outputs via known computer vision systems.
2. Measure the performance of a single subs system to prove both design and economic viability
a. End effector (each) will be capable of picking once every 3-4 seconds
b. End effector (each) will be engineered to reliably perform 3 million-plus actuations
c. Full gripper and subsystem (Harvester Wall with 8 end-effector configuration) will be engineered for sub-half second harvest while moving through an orchard on a platform.
Project Title: Smart Orchards Year 3 + Connectivity
Organization (s): Innov8 ag, Washington State University
PI (s): Mantle, S.; Khot, L.; Bolivar Medina, J.
Total Funding Amount for All Years: $95,500
Length: 1 year
Availability of real-time orchard site information, specifically local weather conditions, soil water and nutrient status, and canopy vigor/tree health, will permit growers to precisely execute timely management decisions and avoid crop losses, thereby enhancing the competitiveness of the Washington’s fruit crops.
Washington apple industry buy-in of approach is anticipated, as we break down existing data siloes. Through data analytics of collected data, we expect: 1) identification of data products (indicators) that better reflect vigor status in the orchard block, better mapping of abiotic crop stressors, and decision aid for water and nutrient management, and 2) increased level of knowledge of data driven technology for practical management execution. The tree fruit industry will benefit from long-term impacts realized by the reduction of use of resources, while improving their production and pack-outs. The industry will also benefit from the independent validation and knowledge of compatible sensing solutions that fit in complete automation concepts to realized smart orchards of the future.
The primary objectives of this effort will be:
- Enable deeper collaboration with smart orchard growers (and interested community) to implement practices based on sensory/imagery-driven insights for better yield outcomes, with a particular focus on irrigation and nutrient optimization.
- Build upon base connectivity capability at Grandview site to enable interaction with data by grower managers/employees/advisors, while also engaging connectivity provider(s) (e.g., T- Mobile, PocketiNet) interest in rural farm connectivity.
- Drive further collaboration across the tech system with additional sensors and field days, led by a small group comprised of the co-PIs and coordinated by WSU Tree Fruit Extension’s Jenny Bolivar.
- Early exploration of intersect between automation & sensor/imagery-driven data collection and analysis, by encouraging automation/robotics-enabled scenarios to include usage & collaboration in one or both orchards.
Project Title: Low-cost, reliable soft arm for automated tree fruit operations
Organization (s): Washington State University
PI (s): Luo, M.; Karkee, M.
Total Funding Amount for All Years: $126,656.00
Length: 1 year
To overcome the current obstacles in automating labor-intensive orchard operations, the long-term goal of this research is to develop a universal robotic system for fruit tree orchard applications utilizing a novel, low-cost, and rapid-acting soft manipulator. The proposed system consists of a set of soft, growing manipulators with a local vision-based controller and a universal adapter for installing various
tools/end effectors, a laser scanner for global canopy characterization, and a self-driving ground mobile platform (e.g.Warthog, Clearpath Robotics, available to WSU team). In this project, the focus will specifically be on automated apple harvesting as a use- case using a single soft, growing manipulator. The effort includes robot design and prototyping, local perception system development, planner and controller development, and simulated lab experiments as well as field experiments with the robot. The success with this use- case will be the first milestone towards developing autonomous systems for various orchard operations including harvesting, pruning, thinning, and pollination.
Project Title: Automated Apple Harvester
Organization (s): Advanced Farm Technologies (AFT)
PI (s): Ferguson, P.; Grossman, M.
Total Funding Amount for All Years: $460,000
Length: 3 years
Advanced Farm Technologies (AFT) intends to develop an automated apple harvester to give apple growers a sustainable solution to deal with labor shortages and rising labor costs. 2022 Development Goals: Build at least one robotic apple harvester that will pick in Washington apple orchards for the duration of the 2022 harvest.
- Build an apple orchard emulator. The emulator is a virtual orchard and robot that can be used to test, improve, and debug the software that will control the harvester. The team has done this for strawberries and has found it to be very useful. Since apple harvest is only three months long, the team must arrive on site ready to work. By building a virtual apple orchard, AFT can verify many elements of the harvester hardware and software months before the harvest season begins. The data scientists and software engineers have already visited WA to take extensive measurements at apple orchards to help AFT build a representative virtual orchard.
- Leverage standard AFT harvester software. The software that runs the vision system and drive system for the strawberry harvester is being extracted and standardized so that it can be applied to apples. Utilizing this proven code base will give AFT a head start on the apple harvester.
- Build a gripper that can pick apples while minimizing drops and fruit damage. Currently gripper testing is being conducted. Different methods of grasping the apple have been tried, and AFT will proceed with a suction cup design. This suction cup design has been initially evaluated for picking thoroughness, fruit bruising mitigation, and efficacy across different varieties.
Project Title: Multi-purpose Robotic System for Orchards
Organization (s): FFRobtics Ltd
PI (s): Kahani, A.; Koster, Y.
Total Funding Amount for All Years: $150,000
Length: 1 year
The following are the project objectives
- Integrate and demonstrate multi-arm harvesting robot to cover entire tree height
- Evaluate the performance of the harvesting robot while in motion
- Improving robot throughout
- Demonstrate integration of the harvesting robot with fruit conveying and bin filling system
Contact:
Paige Beuhler (Administrative Officer): paigeb@treefruitresearch.com, 509 665 8271 ext. 2 Ines Hanrahan (Executive Director): hanrahan@treefruitresearch.com; 509 669 0267
Fruit Matters articles may only be republished with prior author permission © Washington State University. Reprint articles with permission must include: Originally published by Washington State Tree Fruit Extension Fruit Matters at treefruit.wsu.edu and a link to the original article.
