Manual Feed Collision Prediction Using Learned Pulse Waveforms
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In machine tools with manual feed capabilities, predicting the future position of movable parts is challenging due to real-time user operation, leading to excessive alarm generation and reduced work efficiency, as existing methods assume continuous manual handle operation and may incorrectly predict interference with objects.
Innovation Solution
A machine learning device that acquires manual feed state information, including pulse waveforms, and generates a learned model through supervised learning to predict the moving distance of movable parts, reducing false alarm generation by accurately anticipating user operation patterns and interference risks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the future position of the movable part is predicted based on the assumption that the present pulse signal state is maintained, then the interference check can be performed, but excessive alarms are generated and work efficiency declines
Solution Approach 1:
The system performs preliminary learning of user operation patterns before actual manual feed operations. By acquiring pulse waveforms and actual moving distances during learning operations, the system builds a learned model that predicts future positions more accurately, reducing false alarms while maintaining collision prevention capability
Solution Approach 2:
The system incorporates feedback by comparing predicted positions with actual positions and continuously refining the learned model. The prediction unit uses the learned model to predict future positions, and this prediction is fed back into the interference check process, improving accuracy over time while reducing excessive alarms
2Reliability
If the interference check is performed using calculated future position based on pulse signals, then collision prevention is achieved, but false alarms interrupt user operation
Solution Approach 1:
The system performs preliminary learning operations to capture pulse waveforms and corresponding actual moving distances before normal operation. This preliminary data collection enables the learned model to understand the relationship between pulse signals and actual movement, improving prediction accuracy and reducing false alarms that would interrupt user operation
Solution Approach 2:
The system changes from using simple pulse signal integration to using a learned model that incorporates pulse waveform characteristics and actual movement patterns. This parameter change in the prediction method improves accuracy by capturing the nonlinear relationship between manual handle operation and movable part position, thereby reducing false alarms
Data Source
AI summary
To prevent a collision of a movable part without generating an alarm excessively in manual feed. A machine learning device includes: a state observation unit that acquires, as input data, manual feed state information including a manual feed pulse waveform at a time of a manual feed operation in any manual feed operation performed in a machine tool capable of manual feed; a label acquisition unit that acquires label data indicating a distance by which a movable part of the machine tool moved within a predetermined time immediately after the manual feed pulse waveform of the manual feed state information included in the input data; and a learning unit that executes supervised learning by using the input data acquired by the state observation unit and the label data acquired by the label acquisition unit, and generates a learned model.