Related Experiment Video
Updated: Jan 1, 2026
05:47
Simulation of a Scaled Assembly Process with Collaboration of a Robotic Arm and Monitoring through a Vision System for Quality Control
Published on: August 29, 2025
308
An Autonomous Fruit and Vegetable Harvester with a Low-Cost Gripper Using a 3D Sensor
Tan Zhang1, Zhenhai Huang1, Weijie You1
1Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ), Shenzhen University, Shenzhen 510000, China.
Sensors (Basel, Switzerland)
|December 28, 2019
Summary
This study introduces an autonomous robotic system for harvesting crops with peduncles. The system uses a Mask Region-based Convolutional Neural Network (Mask R-CNN) and a novel gripper for efficient and non-damaging fruit harvesting.
Area of Science:
- Agricultural Robotics
- Computer Vision
- Mechanical Engineering
Background:
- Automated harvesting is essential for agricultural efficiency.
- Existing systems struggle with unstructured environments and diverse crops.
Purpose of the Study:
- To develop an autonomous system for harvesting crops with peduncles.
- To enable robots to efficiently harvest fruits and vegetables in real-world conditions.
Main Methods:
- Adapted Mask Region-based Convolutional Neural Network (Mask R-CNN) for fruit detection and bounding box generation.
- Developed a novel gripper for simultaneous clamping and cutting of crop peduncles.
- Integrated a geometric approach to determine precise peduncle cutting points.
Main Results:
- The system successfully identified cutting points using fruit bounding boxes.
- The novel gripper demonstrated effective clamping and cutting without damaging the crop flesh.
- Experimental evaluation with a robotic manipulator confirmed the system's effectiveness in laboratory settings.
Conclusions:
- The proposed autonomous system offers a robust solution for harvesting crops with peduncles.
- The integration of advanced object detection and a specialized gripper enhances harvesting efficiency and precision.
- This technology has the potential to significantly advance automated agricultural practices.
