Neural Network Motion Control for Variable Setpoint Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In semiconductor manufacturing, existing motion control systems face inaccuracies due to non-repetitive setpoints and disturbance forces, making it challenging to achieve precise component movement, especially in lithographic apparatuses where features are increasingly smaller and require sophisticated control techniques.
Innovation Solution
Implementing a system that uses a trained artificial neural network to determine control outputs for component movement, allowing for accurate control regardless of whether the input falls within the training data, thereby enhancing component movement accuracy and reducing process setup time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional motion control systems are used with non-repetitive setpoints and disturbance forces, then the system can handle varied motion requirements, but component movement accuracy deteriorates
Solution Approach 1:
The system performs preliminary measurements of disturbance forces during component movement and stores this data for later use. By measuring and storing disturbance force data in advance, the system prepares compensation information that can be applied in subsequent operations, enabling accurate control even with non-repetitive motion patterns
Solution Approach 2:
The system implements feedback by measuring actual disturbance forces during component movement and using this measured data to generate compensation signals. The disturbance force measurement unit continuously monitors forces acting on the component, and this feedback information is used to adjust control outputs for improved accuracy
2Manufacturing precision
If sophisticated fine-tuning steps and tight control loops are applied, then manufacturing precision improves, but device complexity increases
Solution Approach 1:
The system introduces a disturbance force measurement unit as an intermediary component that directly measures the forces acting on the motion control system. This measurement intermediary provides accurate disturbance force data that simplifies the control algorithm by providing direct feedback about actual disturbances, reducing the need for complex modeling and compensation calculations
Solution Approach 2:
The system replaces complex mechanical control mechanisms with a measurement-based control approach. Instead of using sophisticated mechanical fine-tuning mechanisms and tight control loops, the system uses disturbance force measurements to generate compensation signals, substituting mechanical complexity with measurement and calculation
Data Source
AI summary
Variable setpoints and/or other factors may limit iterative learning control for moving components of an apparatus. The present disclosure describes a processor configured to control movement of a component of an apparatus with at least one prescribed movement. The processor is configured to receive a control input such as and/or including a variable setpoint. The control input indicates the at least one prescribed movement for the component. The processor is configured to determine, with a trained artificial neural network, based on the control input, a feedforward output for the component. The artificial neural network is pretrained with a training data set such that the artificial neural network determines the output regardless of whether or not the control input falls outside the training data set. The processor controls the component based on at least the output.