Development of a Machine Vision-Based Robotic Apple Crop Load Management System at Bud Stage

Commodities

• Fruits: apples

Practices

• Crop Production: crop improvement and selection
• Education and Training: decision support system, extension

Apple crop load management is critical for balancing high yield
and quality fruit production while ensuring sustainable tree
health. Artificial Spur Extinction (ASE) is a new crop management
technique that reduces the floral bud density of the tree canopy.
This project aims to optimize the ASE process in apple orchards
through advanced vision-based robotics, enhancing fruit quality
and ensuring regular annual bearing. This technique employs
automated, precise bud thinning based on the branch
cross-sectional area (BCA) to achieve the desired fruit size. Our
study follows three primary objectives: Firstly, developing a
vision-based system for assessing bud density at the tree and
branch levels enables accurate thinning decisions. Second, a
decision support system for bud thinning will be developed based
on the bud density and distributions at the branch and tree
levels. Lastly, the robotic thinning system undergoes integration
and evaluation through comprehensive field testing. Key to the
success of this study is the engagement with the local
agricultural community through a comprehensive outreach strategy.
The technology will be demonstrated on field days targeted at
local apple growers, and the findings and methodologies will be
shared through extension talks and research publications. This
approach facilitates the direct transfer of technology and
encourages feedback for further refinement. The expected outcome
is a proof of concept to showcase the feasibility and benefits of
robotic bud thinning, potentially advancing apple orchard
management and significantly boosting profitability through
improved crop quality and yield.

This proposed research aims to develop a robotic bud thinning
system for early crop load management in orchards, considering
branches and tree levels for good-quality apples. The project
will be structured around three key objectives:

Objective #1: Develop Deep Learning-Based
Algorithms for Apple Bud Detection and Localization in Orchard
Environments

Accurate detection and localization of apple buds within diverse
orchard environments are foundational to precision agriculture
practices, particularly in the context of bud thinning. This
objective addresses a critical gap in current agricultural
technology by developing sophisticated deep-learning algorithms
that analyze visual data for bud density and specific locations.
These algorithms are about identifying the buds and understanding
the intricate dynamics of crop load at both branch and tree
levels. The precision and effectiveness of bud thinning, directly
influenced by the accuracy of these detections, can significantly
affect the fruit quality and quantity, making this technology
pivotal for sustainable and profitable orchard management.

Objective #2: Create a Decision Support System
for Bud Thinning Based on the Bud Density and Distributions at
the Branch and Tree Levels

Creating a decision support system (DSS) for bud thinning based
on a comprehensive analysis of bud density and distribution
represents a leap towards intelligent crop management. Such a
system moves beyond the conventional one-size-fits-all approach
to thinning, offering instead a nuanced, data-driven strategy
that considers the unique characteristics of each tree and
branch. This approach optimizes resource use and ensures that
each action taken contributes to the overall health and
productivity of the orchard. The DSS shifts from manual to
data-driven decision-making, offering insights that boost
thinning efficiency, cut labor costs, and enhance crop yield and
quality.

Objective #3: Design and Test a Robotic
End-Effector for Effective Bud Removal for Early Apple Crop Load
Management

Designing and testing a robotic end-effector for efficient bud
removal tackles one of the most labor-intensive and
skill-dependent tasks in orchard management. Introducing such a
device is not merely an advancement in mechanical design but a
fundamental shift towards fully automated, precision agriculture.
The effectiveness of bud thinning, crucial for controlling crop
load and ensuring the development of high-quality fruit, depends
significantly on the precision and gentleness of the removal
process. A robotic end-effector that can quickly, accurately, and
safely remove unwanted buds without damaging the tree opens the
door to unprecedented efficiency and effectiveness in orchard
management.