Related Experiment Video
Updated: Jul 4, 2025
08:02
Robotic Sensing and Stimuli Provision for Guided Plant Growth
Published on: July 1, 2019
8.0K
Intelligent robotics harvesting system process for fruits grasping prediction
K M Alaaudeen1, Shitharth Selvarajan2,3, Hariprasath Manoharan4
1Department of Computer Science and Engineering, Grace College of Engineering, Mullakkadu, Thoothukoodi, India.
Scientific Reports
|February 2, 2024
Summary
This study introduces an AI-powered vision system for robotic apple harvesting. The system achieves over 95% fruit identification accuracy, enabling precise robotic arm control for efficient automated harvesting.
Area of Science:
- Computer Vision
- Robotics
- Artificial Intelligence
- Agricultural Technology
Background:
- Manual harvesting is labor-intensive and costly.
- Automated harvesting systems require sophisticated fruit recognition and precise manipulation.
- Existing vision systems may lack efficiency and accuracy in complex environments.
Purpose of the Study:
- To develop and evaluate a deep learning-based image processing system for autonomous apple harvesting.
- To integrate fruit recognition, instance segmentation, and grasping point prediction for robotic arms.
- To assess the system's performance in diverse laboratory and field conditions.
Main Methods:
- A lightweight, one-step detection network was employed for fruit recognition.
- Computer vision techniques analyzed fruit class and determined optimal grasping positions from RGB images.
- The vision system was integrated with a robotic arm for autonomous harvesting experiments.
Main Results:
- The system achieved over 95% accuracy in fruit identification.
- Post-prediction processes required less than 12% reattempts.
- The robotic harvesting system demonstrated effective and precise control in both indoor and outdoor experiments.
Conclusions:
- The proposed deep learning-based vision system effectively controls robotic harvesting operations.
- High identification accuracy and efficient grasping point prediction are crucial for successful automated harvesting.
- The system shows significant potential for improving efficiency and reducing costs in fruit harvesting.
