NC Corner Path Segmentation Using Reinforcement Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In numerical control systems, acceleration/deceleration control in corner portions leads to inward turning and deviations from the original machining path, resulting in reduced machining accuracy and increased time.
Innovation Solution
A machine learning device that analyzes machining programs and replaces two-block commands with m or more blocks, using reinforcement learning to adjust coordinate values and optimize the machining path, thereby reducing inward turning and improving accuracy while decreasing machining time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If acceleration/deceleration control is performed in corner portions, then machining accuracy is improved, but machining time increases
Solution Approach 1:
The patent segments the corner portion machining into multiple blocks (m or more blocks instead of two blocks) with divided coordinate values. By dividing the corner movement into multiple intermediate steps with adjusted coordinate values, the system can apply acceleration/deceleration control at each segment, improving machining accuracy while optimizing the overall machining time through learned coordination of these segments.
2Manufacturing precision
If coordinate values are adjusted to reduce inward turning, then machining accuracy is improved, but the complexity of control increases
Solution Approach 1:
The patent implements reinforcement learning where the machine learning device receives feedback in the form of reward values based on machining accuracy and machining time. The device learns optimal coordinate value adjustments by continuously receiving feedback on the outcomes of previous adjustments, automatically optimizing the control parameters without requiring complex manual programming or adjustment procedures.
Data Source
AI summary
A machine learning device performs machine learning on a numerical control device which, when a first command including a corner portion, composed of two blocks in the machining program, generates a second command in which the two blocks are replaced with m or more blocks. The machine learning device comprises: a state information acquisition unit for acquiring state information including the first command, coordinate values of each block in the m or more blocks, and location information of the machining path and the machining time; an action information output unit for outputting action information; a reward output unit for outputting a reward value based on the inward turning amount in the corner portion; and a value function updating unit for updating a value function based on the value of the reward outputted from the reward output unit, the state information and the action information.