Development of Orchard Robot Control System Using ROS 2

<b><sc>Abstract.</sc></b> The Robotics Vision Lab at Northwest Nazarene University has developed the Orchard Robot (OrBot), an innovative solution to assist farmers in orchard management. OrBot is equipped with an articulated arm, a machine vision system, a customizable end effector, a mobile platform, and an onboard computer. The initial control platform for OrBot was developed using Matlab and its associated toolboxes. OrBot was first tested for apple harvesting in a commercial orchard, achieving a success rate of approximately 80% with a cycle time of 15 seconds. The objective of this research was to enhance OrBot's control system by transitioning to ROS 2. All the components of OrBot were assigned nodes: the vision system, manipulator, end effector, mobile tank, and image processing. A master node was created to coordinate all these different nodes to perform the harvesting task. OrBot was tested indoors using artificial trees and artificial apple fruits. The artificial fruits were positioned randomly on the tree and OrBot had a 100% success rate on all the trials. OrBot was then tested in a commercial orchard last year during the harvesting season of Pink Lady apples. OrBot had a success rate of about 75% and an improved cycle time of about 10 seconds.