Edge ML Processing for High-Speed Workpiece Portioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
High-speed portioning machines struggle to accurately process workpieces with varying shapes, dimensions, weights, densities, colors, and textures due to limitations in sensor data processing and machine learning capabilities.
Innovation Solution
Implementing a local high-power computing device, or edge computing device, that receives sensor data from workpieces, processes it, and outputs information to optimize machine processing, including using machine learning models for image processing and feature recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-speed portioning machines use standard sensor data processing, then processing speed is maintained, but accuracy of workpiece analysis deteriorates due to variations in shapes, dimensions, weights, densities, colors, and textures
Solution Approach 1:
The system segments the data processing task by introducing an edge computing device that handles complex machine learning model execution separately from the main high-speed portioning machine. This allows the main machine to maintain its high processing speed while the edge device performs detailed image processing and feature recognition to improve measurement accuracy.
Solution Approach 2:
An edge computing device is introduced as an intermediary between the sensor/input and the main portioning machine. This intermediary device receives sensor data, executes machine learning models for detailed analysis, and returns processed information to the main machine, thereby improving accuracy without compromising the main machine's processing speed.
2Measurement precision
If high-power computing devices are added to improve processing capabilities, then analysis accuracy improves, but device complexity increases
Solution Approach 1:
The edge computing device serves as a specialized intermediary that handles complex computing tasks. By separating the high-power computing requirements from the main portioning machine, the system improves data processing accuracy while containing complexity in a dedicated, manageable component rather than complicating the entire system.
Solution Approach 2:
The complex machine learning processing capabilities are extracted from the main portioning machine and placed in a separate edge computing device. This extraction allows the main machine to remain relatively simple while still benefiting from advanced processing capabilities when needed.
3Productivity
If machine learning models are executed locally on the portioning machine, then processing speed is maintained, but computational power requirements exceed available resources
Solution Approach 1:
The edge computing device acts as an intermediary that provides the necessary computational power for machine learning model execution. It receives data from the high-speed portioning machine, performs computationally intensive processing, and returns results in time to maintain the overall processing speed of the system.
Solution Approach 2:
The computational tasks are segmented between the main portioning machine (which handles high-speed control) and the edge computing device (which handles intensive machine learning computations). This segmentation allows each component to operate within its optimal performance and power constraints while achieving the desired overall processing speed.
Data Source
AI summary
A computer-implemented method of optimizing machine processing of a workpiece may include receiving, by a computing device, at least one sensor input regarding a workpiece; performing, by a computing device, pre-processing of the at least one sensor input for at least one of efficient transfer to another computing device and optimal use in one or more machine learning models; executing, by a computing device, one or more machine learning models to output requested information regarding the workpiece based on data in the at least one sensor input; processing, by a computing device, the output; and controlling at least one aspect of the machine processing of the workpiece, by a computing device, in response to the processed output.