Student Spotlight on Kittiphum Pawikhum: Bud Control for Better Apples: The Future of Fruit Farming 

Producing high-quality apples requires effective crop load management, which involves adjusting the number of apple buds during winter. This process is typically performed by skilled workers who manually remove buds. It is crucial for overcoming nutritional limitations and ensuring the trees produce fruit that meets high standards in terms of size, color, taste, and overall marketability. In Pennsylvania, with its 52,000 acres of orchards as of 2022, careful winter management is essential for achieving this desired fruit grade.

Kittiphum Pawikhhum, a Ph.D. student in Agricultural and Biological Engineering, is working to automate this process. "Adjusting bud counts, once a labor-intensive and time-consuming task, can now be much easier with robotics and AI. My research is helping to fully automate thinning and efficiently remove extra apple buds." Kittiphum explains. His research focuses on using robots to control bud numbers, which directly impacts fruit production. Traditionally, deciding which buds to remove has been a challenging task that relied heavily on expert judgment. Kittiphum’s research aims to combine cameras, computer vision, and AI to identify the buds that need to be removed. The system autonomously considers factors like weather, apple variety, and bud density to make these decisions.

During winter, apple trees develop small shoots at the ends of branches, which grow into buds. These buds eventually become flowers and then fruit. If too many buds form, the limited nutrients can lead to lower-grade apples. Controlling the number of buds is crucial for producing premium apples, even if it means producing fewer. Kittiphum’s research enables this process to be done cost-effectively using a compact system with a camera mounted on a moving platform, with most functions controlled by software. This approach makes it affordable and accessible for many farmers.

Kittiphum’s research also automates bud removal using robots. This requires precise localization of the detected buds and accurate control of the robot to remove tiny buds without damaging the branches. Since this task must be performed in environments that are not controlled or predictable, sophisticated systems are necessary to manage various variables. Additionally, to make this approach practical, it is important to use minimal hardware to keep costs down while ensuring effectiveness.

Kittiphum’s research is pioneering the integration of natural agricultural processes with modern tools. By exploring how precise bud removal can be automated in complex orchard environments, Kittiphum and his team are pushing the boundaries of agricultural robotics. Unlike traditional chemical methods that can harm soil health and lack selectivity, Kittiphum’s approach uses computer vision, AI, and robotics to create a more sustainable and precise solution. This technology not only preserves the natural ecosystem but also ensures accuracy in bud removal, leading to higher fruit quality and consistency.

By leveraging these innovative methods, Kittiphum is setting a new standard in precision agriculture that enhances both efficiency and sustainability. His work not only addresses the challenge of automating bud removal but also lays the foundation for broader applications in agricultural robotics. Kittiphum’s commitment to bridging the gap between natural processes and robotic applications reflects his dedication to innovation in this field. Through their efforts, Kittiphum and his team at Penn State are laying the groundwork for a future where advanced technology seamlessly integrates with nature’s processes, revolutionizing the tree fruit industry.

__________

Kittiphum Pawikhhum is a Ph.D. student advised by Professor Paul Heinemann, and Long He in the Agricultural and Biological Engineering Department at Penn State. His research is supported by funding from the United States Department of Agriculture (USDA)’s National Institute of Food and Agriculture (NIFA) Federal Appropriations under Project #PEN04822 and Accession #7005925, and USDA-NIFA SCRI program #2020-51181-32197,  and Royal Thai Government Scholarship.

This article was written by Donghyeon Lee, an INSECT NET Fellow and Ph.D. student advised by Professor Bo Cheng in the Mechanical Engineering Department at Penn State. It was prepared as part of the INSECT NET Science Communication workshop series coordinated by Drs. Christina Grozinger and Natalie Boyle in summer 2024.