3D Vision Mushroom Harvester for Selective Low-Damage Picking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current mushroom harvesting methods, particularly in Agaricus bisporus cultivation, face challenges such as damage to mushrooms due to automated picking devices, disruption of growing environments, and inefficiencies in selective harvesting, which are exacerbated by the limited space in standard grow bed systems and dynamic properties of mushroom growth.
Innovation Solution
An automated harvester system with a vision system using 3D scanners and a picking system with grippers that mimic human picking motions, capable of navigating existing grow bed infrastructure to selectively harvest mushrooms with minimal damage, while maintaining optimal growing conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If standard automated picking devices are used, then harvesting efficiency is improved, but mushroom damage increases
Solution Approach 1:
The patent replaces traditional mechanical picking devices with a vision-guided robotic system that uses 3D scanners to detect mushroom properties and controlled grippers to perform selective harvesting. This substitution enables precise measurement and gentler handling, reducing mushroom damage while maintaining harvesting efficiency.
Solution Approach 2:
The system creates a digital replica of the mushroom bed using 3D scanning technology, generating point cloud data that represents the physical environment. This digital copy allows for virtual analysis and planning of picking operations, enabling precise identification of optimal harvest points without physical contact until the final picking action.
2Area of stationary object
If standard grow bed systems are used, then space utilization is improved, but automation compatibility deteriorates
Solution Approach 1:
The patent employs a mobile robotic platform that can dynamically navigate through the grow bed system. The robot's ability to move and reposition itself allows it to access mushrooms in tightly spaced configurations without requiring fixed infrastructure modifications, adapting to the dynamic constraints of standard grow beds.
Solution Approach 2:
The system transitions from 2D camera-based detection to 3D scanning using point cloud technology. This dimensional enhancement allows the robot to perceive depth, height, and spatial relationships accurately, enabling it to operate effectively in the limited vertical and horizontal spaces of standard grow beds while maintaining automation capabilities.
3Reliability
If manual harvesting is used, then mushroom quality is improved, but labor intensity increases
Solution Approach 1:
The robotic system performs autonomous operation, independently detecting mushrooms, determining optimal harvest timing, and executing picking actions without continuous human intervention. The vision system continuously monitors mushroom growth and automatically identifies ready-to-harvest specimens, reducing labor input while maintaining quality standards.
Solution Approach 2:
The system implements continuous feedback loops where the vision system monitors mushroom growth in real-time, updates the digital model, and adjusts picking decisions based on current conditions. This feedback mechanism ensures consistent quality assessment and timing, matching or exceeding manual harvesting reliability while eliminating labor-intensive monitoring.
4Productivity
If selective harvesting is implemented, then crop yield is improved, but detection precision requirements increase
Solution Approach 1:
The patent replaces human visual inspection with 3D scanning technology that captures precise geometric data of each mushroom. The point cloud representation enables accurate measurement of mushroom size, shape, and position, providing the detection precision needed for selective harvesting decisions that maximize crop yield.
Solution Approach 2:
The system performs preliminary detection and classification of mushrooms before the actual picking action. The vision system scans and identifies target mushrooms in advance, allowing the control system to plan optimal picking sequences and adjust harvesting strategies based on the complete view of the mushroom bed, thereby improving overall crop yield.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Provided are a system, method(s), and apparatus for automatically harvesting mushrooms from a mushroom bed. The system, in one implementation, may be referred to herein as an "automated harvester", having at least an apparatus/frame/body/structure for supporting and positioning the harvester on a mushroom bed, a vision system for scanning and identifying mushrooms in the mushroom bed, a picking system for harvesting the mushrooms from the bed, and a control system for directing the picking system according to data acquired by the vision system. Various other components, sub-systems, and connected systems may also be integrated into or coupled to the automated harvester.