Adaptive Optical Sorting Adjusting Evaluation Parameters for Real-Time Decisions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Optical sorting systems face challenges in achieving real-time sorting decisions due to high computational demands and varying object densities in material flows, leading to potential incorrect sorting and reduced performance.
Innovation Solution
An adaptive sorting system that adjusts calculation times and accuracy levels based on occupancy parameters, using image processing algorithms to ensure timely sorting decisions by setting evaluation parameters such as accuracy, calculation time, and repetition frequency, allowing for incremental refinement of sorting decisions within allocated time budgets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If the distance between the image recording unit and sorting unit is reduced to improve sorting precision, then the position prediction of objects becomes more accurate and by-catch is reduced, but the evaluation unit has less time to process image data and make sorting decisions
Solution Approach 1:
The system dynamically adjusts evaluation parameters (accuracy, calculation time, repetition frequency) based on occupancy parameters detected in real-time. When occupancy is high, the system reduces evaluation accuracy or calculation time to ensure real-time decisions are made. This dynamic adaptation allows the system to maintain optimal performance across varying material flow conditions while respecting the fixed distance constraint between sensor and sorting unit.
2Manufacturing precision
If the accuracy and calculation time for image evaluation are increased to improve sorting quality, then more precise sorting decisions can be made, but the real-time capability of the system deteriorates and sorting decisions may be missed
Solution Approach 1:
The system changes evaluation parameters (accuracy, calculation time, repetition frequency) based on detected occupancy parameters. The control unit adjusts these parameters dynamically to optimize the balance between sorting quality and real-time performance. This allows the system to adapt its computational effort to the actual processing needs while maintaining real-time capability under varying load conditions.
Solution Approach 2:
The system performs partial evaluation when time is constrained, making do with reduced accuracy or fewer calculation iterations rather than attempting complete evaluation. This ensures that some sorting decision is always made in real-time, even if not all computational analyses are fully completed. The system accepts partial results when necessary to maintain real-time operation.
3Productivity
If the material flow occupancy density is increased to improve economic efficiency, then more objects can be sorted per unit time, but object agglomerates form that are difficult to separate algorithmically and increase computational complexity
Solution Approach 1:
The system performs preliminary detection of occupancy parameters before full evaluation begins. Based on this preliminary information, it pre-adjusts evaluation parameters to anticipate computational challenges. This preliminary action allows the system to prepare appropriate computational strategies in advance, preventing excessive complexity from overwhelming the real-time processing capability when high-density material flows are detected.
4Productivity
If the evaluation accuracy is reduced to maintain real-time processing speed, then sorting decisions can be made faster, but the sorting quality and precision deteriorate
Solution Approach 1:
The system dynamically adapts evaluation accuracy based on real-time occupancy conditions rather than using a fixed accuracy level. When occupancy is low, higher accuracy is used to maximize sorting quality. When occupancy is high, accuracy is reduced to maintain real-time processing. This dynamic approach optimizes the trade-off between speed and accuracy according to actual operational conditions.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Ensures a sorting decision is made for each object in real-time, improving sorting performance and quality by minimizing latency and reducing by-catch, while maintaining economic efficiency and flexibility across different sorting tasks.
Implementation Method 1
an image recording unit (1) with which the material flow (M) can be optically detected and with which image data (4) can be generated
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a sorting device with which objects (0) of a material flow (M) are sorted, comprising an image-capturing unit (1) with which the material flow (M) is optically captured and image data (4) of same (M) is generated, an evaluation unit (2) with which objects (0) in the material flow (M) are identified and classified, and a sorting unit (3) with which classified objects (0) of the material flow (M) are sorted, characterised in that one or more configuration parameters (5) characterising the material flow (M) in terms of the configuration of the objects (0) therein is determined with the evaluation unit (2) from the generated image data (4), and in that one or more evaluation parameters (6, 8) controlling the identification and classification of the objects (0) via the evaluation unit (2) is/are used on the basis of the determined configuration parameter/s (5).