Mushroom Trimming and Sorting Robotics for Selective Stem Cleaning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing commercial mushroom cultivation and harvest systems rely heavily on manual labor, which is costly and inefficient, and existing automated systems do not optimize yield and effectiveness, particularly in trimming and sorting mushrooms to maximize commercial value.
Innovation Solution
A mushroom processing system utilizing machine-learning controlled robotics and computer vision to automate stem trimming and sorting, optimizing stem length and quality based on factors like cap size, cleanliness, and gill opening, while minimizing soil and defects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual labor is used to harvest and trim mushrooms, then flexibility and adaptability in handling are improved, but labor costs and time consumption increase significantly
Solution Approach 1:
The patent replaces manual mechanical operations with an automated robotic system equipped with computer vision and machine learning. The robotic arm with adaptive gripper performs harvesting, trimming, and sorting operations that were traditionally done manually, thereby increasing productivity while maintaining operational flexibility through intelligent control algorithms.
Solution Approach 2:
The system incorporates automated decision-making capabilities where the machine learning model determines optimal trimming lengths, sorting categories, and harvesting timing without human intervention. The robotic system serves itself by autonomously navigating the mushroom bed, identifying targets, and executing operations based on real-time visual data analysis.
2Productivity
If automated harvesting systems are implemented, then productivity and harvesting speed are improved, but manufacturing complexity and system cost increase
Solution Approach 1:
The robotic system is designed to perform multiple functions including harvesting, stem trimming, quality assessment, and sorting within a single integrated platform. This multi-functionality reduces the need for separate automated systems for each operation, thereby managing complexity while maintaining high productivity across multiple process stages.
Solution Approach 2:
The system dynamically adjusts operational parameters such as gripper force, cutting depth, and sorting thresholds based on real-time mushroom characteristics detected by computer vision. This adaptability allows the system to handle varying mushroom sizes, shapes, and firmness without requiring complex reconfiguration, thus managing system complexity while maintaining high productivity.
3Manufacturing precision
If extensive trimming is performed to remove soil and defects, then product quality and commercial value are improved, but loss of substance and yield decrease
Solution Approach 1:
The system uses computer vision to continuously monitor mushroom stem conditions and provides feedback to the trimming mechanism. The machine learning model analyzes the degree of soil contamination and defect locations, then adjusts trimming depth in real-time to remove only the necessary portions, minimizing weight loss while ensuring product cleanliness and quality standards are met.
Solution Approach 2:
The trimming operation is applied locally and selectively to specific portions of the mushroom stem that require cleaning, rather than uniform trimming across all mushrooms. The system identifies and targets only the contaminated or defective areas, preserving as much edible mushroom tissue as possible while achieving the required cleanliness standard.
4Productivity
If frequent harvesting is performed to prevent overgrowth, then productivity and total yield are improved, but loss of time and operational complexity increase
Solution Approach 1:
The robotic harvesting system operates continuously without the breaks and transitions required by manual harvesting crews. The automated system can rapidly move between mushroom beds, harvest, trim, and sort without interruption, enabling more frequent harvesting cycles and maximizing total yield from each flush while minimizing idle time between operations.
Solution Approach 2:
The system performs preliminary assessment of mushroom readiness using computer vision before harvesting begins. The machine learning model predicts optimal harvesting timing based on growth rate analysis, allowing the system to prepare for and execute harvesting at the precise moment when mushrooms are ready, thereby maximizing yield without requiring excessive waiting time or frequent unnecessary trips to the mushroom beds.
Data Source
AI summary
A mushroom trimming and sorting system comprises a stem trimming device, a positioning system, an imager set, and a controller operable to control the positioning system to move a single mushroom into view of the imager set, to generate an image set of the mushroom, and to process the image set (and optionally other sensor data) to determine mushroom properties of the mushroom. The positioning system may include a SCARA robot with suction-type end effector and single-mushroom elevator with gripping mandibles. The processing may employ a trained machine-learning model. The mushroom properties may include a length of an end portion of the stem bearing soil, and a sorting category of the mushroom based on parameters of the cap including degree of gill opening. The end portion is them trimmed to eliminate soil and other defects, and sorting the mushroom into one of a plurality of receptacles.