Development of a Robotic Pruning System for Sustainable Apple Production

Project Overview

GNE19-225
Project Type: Graduate Student
Funds awarded in 2019: $15,000.00
Projected End Date: 07/31/2021
Grant Recipient: The Pennsylvania State University
Region: Northeast
State: Pennsylvania
Graduate Student:
Faculty Advisor:
Long He
Pennsylvania State University

Information Products

Commodities

  • Fruits: apples

Practices

  • Crop Production: Pruning

    Proposal abstract:

    Despite the success of automation in row crops, production operations of fruit-bearing trees still largely depend on manual labor. Pruning of apple trees requires 80-120 working hours of skilled labor per hectare accounting for 20% of the total production cost. The limited labor pool and associated costs have raised concern among apple growers. The study aimed at developing an end effector and vision system for robotic tree pruning to promote economical and sustainable apple production systems. The proposed project is comprised of developing a robotic pruning end effector for cutting branches, and a machine vision system to detect the tree branches for pruning location identification. An end-effector will be designed by considering the spatial and horticultural requirements of the apple trees. The amount of force required to cut the branches will be recorded for the end-effector design. The end-effector will be integrated with a robotic manipulator. The simulation of the integrated system space utilization for branch accessibility will also be performed and the field tests will be conducted to validate the simulated results. A vision system will be developed for 3D reconstruction of the canopy. Pruning parameters will be optimized by incorporating the generally adopted pruning rules, and the pruning location will be identified with the machine vision system. The performance will be assessed in terms of branch identification accuracy, pruning branch proportion, false negative branches, and false positive branches. The successful completion of this project will help to develop an automated pruning system for apple trees.

    Project objectives from proposal:

    The primary goal of this project is the development of a robotic system for apple tree pruning using a robotic end-effector and machine vision system. The project will be focused on the following objectives:

    1. Development of an effective robotic pruning end-effector The end effector will be developed by considering the maneuvering capabilities to cut the branches of all orientations and minimum spatial requirements to freely move in a limited workspace. The field test for pruning force measurement will be conducted which is essential for the selection of the end effector components. The design optimization and orientation tracking of the end effector will also be performed. It is expected that the end effector design will be able to produce smooth branch cuts, utilize the minimum workspace, and cut the branches of all orientation.

    2. Integration of a robotic pruning system and simulation of spatial requirements for branch accessibility The developed end effector will be integrated with two different robotic arms i.e. three axes linear motion arm and a six degrees of freedom robotic arm. The simulation of the integrated system in a virtual environment will be performed to estimate the space requirements and branch accessibility of both integrated systems. It is expected that the simulation study for branch accessibility will help in developing a collision-free trajectory and path planning for the end effector to reach the target position.

    3. Development of a vision system for tree branch detection and optimization of pruning parameters The image acquisition system will be developed to acquire the visual information of the tree canopy, and an image processing algorithm for the identification of branches will be developed. To evaluate the developed machine vision system, a pruning rule of removing branches based on a defined limb-trunk ratio threshold will be used. It is expected that the vision system will be able to detect and identify the base diameter of all branches in a tree and provide the cutting location for the integrated robotic pruning system.

    Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.