Component Mounting Machine Adaptive Retry Threshold
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Component mounting machines incorrectly determine components as abnormal due to measurement variances in image processing, leading to unnecessary discarding of valid components and increased cycle time from repeated image processing attempts.
Innovation Solution
A component mounting machine with a control section that compares shape data deviation to a retry determination value, discarding components exceeding this value and retrying image processing if within it, with the retry determination value adjusted based on measurement variance to minimize incorrect determinations and reduce cycle time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If retry image processing is performed when deviation exceeds allowable value, then incorrect discarding of components is reduced, but cycle time increases and productivity decreases
Solution Approach 1:
The patent applies parameter changes by introducing a dynamic threshold value that adjusts based on the number of retry attempts. The threshold is set higher for initial measurements and becomes progressively stricter with each retry, allowing the system to accommodate measurement variance while preventing infinite loops. This resolves the contradiction by enabling reliable component determination through adaptive threshold adjustment without excessively increasing cycle time.
Solution Approach 2:
The patent implements dynamics by making the determination threshold dynamic rather than static. The threshold value changes based on the measurement attempt number, creating an adaptive system that responds to the actual measurement conditions. This dynamic approach allows the system to be more lenient when measurement variance is expected and stricter when confidence increases, thereby improving reliability without permanently extending cycle time.
2Measurement precision
If image processing parameters are changed and retry processing is performed, then measurement variance may be reduced, but components with actual deviation exceeding allowable amount may be incorrectly determined as normal
Solution Approach 1:
The patent applies parameter changes in a controlled manner by adjusting the determination threshold based on the measurement attempt number rather than changing image processing parameters themselves. This approach reduces measurement variance effects while maintaining product quality because the threshold adjustment is systematic and predictable, preventing both types of errors mentioned in the contradiction.
Solution Approach 2:
The patent implements feedback by using the measurement attempt number and previous measurement results to adjust the determination threshold for subsequent measurements. This feedback mechanism allows the system to learn from previous measurements and adjust its criteria accordingly, improving measurement precision while maintaining reliability through iterative refinement rather than arbitrary parameter changes.
3Reliability
If multiple retry image processing attempts are performed for all components exceeding allowable deviation, then incorrect discarding is reduced, but image processing time increases significantly
Solution Approach 1:
The patent applies partial action by performing retry image processing selectively rather than universally. The system performs retries only when the deviation falls within the dynamic threshold range, and the number of retries is limited based on the threshold criteria. This partial approach reduces incorrect discarding for borderline cases while avoiding unnecessary processing time for clearly defective components.
Solution Approach 2:
The patent uses parameter changes to control the extent of retry processing. By adjusting the determination threshold based on the attempt number and comparing against dynamically calculated values, the system automatically limits the number and scope of retries. This parameter-driven approach ensures that retry processing is performed only when beneficial, balancing reliability improvement with time loss minimization.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A component mounting machine including: a camera (32) configured to capture an image of a component held by a suction nozzle (25); and an image processing section configured to process the image captured by the camera, recognize the shape of the component, and measure shape data of the component; wherein, when the deviation amount of the measurement value of the shape data of the component exceeds the allowable value, the deviation amount of the measurement value of the shape data of the component is compared to a retry determination that is larger than the allowance value, and if the deviation amount of the measurement value of the shape data of the component exceeds the retry determination value, the component is determined to be abnormal and is discarded to a specified discard location. In contrast, if the deviation amount of the measurement value of the component shape data is equal to or less than the retry determination value, retry imaging processing is performed in which the camera captures an image of the component held by the suction nozzle again, the component shape data is remeasured, and it is determined whether the component is normal based on whether the remeasured value is equal to or less than the allowance value.