Robotics in high tunnel agriculture: perception-driven harvesting and crop monitoring in unstructured environments

Shah, Poojan Kalpeshbhai

Permalink

https://hdl.handle.net/2142/129970 Copy

Description

Title

Robotics in high tunnel agriculture: perception-driven harvesting and crop monitoring in unstructured environments

Author(s)

Shah, Poojan Kalpeshbhai

Issue Date

2025-07-21

Director of Research (if dissertation) or Advisor (if thesis)

Krishnan, Girish

Department of Study

Mechanical Sci & Engineering

Discipline

Mechanical Engineering

Degree Granting Institution

University of Illinois Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• Agricultural Robotics
• Robotic harvesting
• Autonomous navigation in agriculture
• Eye-in-hand camera systems
• Visual servoing
• Robotic pest inspection

Abstract

Due to labor shortages in specialty crop industries, a need for robotic automation to increase agricultural efficiency and productivity has arisen. Previous manipulation systems harvest well in uncluttered and structured environments. High tunnel environments are more compact and cluttered in nature, requiring a rethinking of the large form factor systems and grippers. We propose a novel co-designed framework incorporating a global detection camera and a local eye-in-hand camera that demonstrates precise localization of small fruits via closed-loop visual feedback and reliable error handling. The system is evaluated in both lab and high tunnel environments for berry harvesting. Second, we extend this framework to DetectCluster2Cut for peduncle-based cluster harvesting of cherry tomatoes, introducing monocular-based depth filtering and peduncle angle estimation. Finally, we propose Nav2Inspect—a navigation and inspection pipeline that integrates CropFollow++ row-following, cluster detection, multi-view close-up imaging, and a fruit-surface pest detection model including pest damages from stink bugs, stilt bugs, and caterpillars. Together, these systems lay a foundation for modular closed-loop perception and manipulation pipelines, offering scalable solutions for real-time in-field harvesting and inspection.

Graduation Semester

2025-08

Type of Resource

Thesis

Handle URL

https://hdl.handle.net/2142/129970

Copyright and License Information

Copyright 2025 Poojan Shah

Owning Collections