Motor Position Control With Delayed Image Feedback Switching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing motor control systems face challenges in achieving fast and accurate positioning due to disturbances, limitations in detectable areas, and long dead times in image processing, leading to destabilization and errors in feedback control systems.
Innovation Solution
A motor control device that combines feedforward and feedback controllers with a machine-end sensor to generate torque commands, using low-pass and high-pass filters to integrate sensor and encoder signals, allowing for accurate positioning by switching control schemes based on sensor availability and minimizing dead time effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image processing is used to detect positional relationship, then positioning accuracy is improved, but control speed deteriorates due to long dead time
Solution Approach 1:
The system performs preliminary actions by capturing images continuously before they are needed for control. The image capturing unit acquires images of the movable component and target point in advance, storing them temporarily so that when control decisions are needed, the images are already available immediately, eliminating the dead time between image capture and analysis.
Solution Approach 2:
The system replaces the traditional sequential process (capture → process → control) with a parallel architecture where image capture continues independently while control calculations use pre-captured images. This substitution of the control workflow eliminates the time delay inherent in sequential processing.
2Speed
If a fixed camera is used to image target points, then imaging speed is improved, but the area of detectable target points deteriorates
Solution Approach 1:
The system makes the imaging system universal by attaching the camera to the movable component itself, allowing it to function as both the moving platform and the observation device. This enables the same camera to image multiple different target points throughout the working area, not just a single fixed area.
Solution Approach 2:
The system transitions from a fixed-position camera (zero-dimensional in space) to a moving camera mounted on the movable component (one-dimensional movement capability). This dimensional change allows the camera to access and image multiple target points across the entire working area while maintaining high imaging speed.
3Adaptability or versatility
If switching control schemes is performed, then adaptability to different detection scenarios is improved, but control stability deteriorates due to value discrepancies
Solution Approach 1:
The system introduces an intermediary switching mechanism that smoothly transitions between control schemes. The switching unit compares the current scheme's detection values with alternative schemes and selects the optimal one, acting as a mediator that prevents abrupt changes and maintains control stability while adapting to different detection scenarios.
Solution Approach 2:
The system makes the control scheme selection dynamic rather than static. The switching unit continuously evaluates which control scheme (encoder-based or image-based) is more appropriate based on current conditions, allowing the system to adapt dynamically while maintaining stability through controlled transitions between schemes.
4Speed
If encoder-based feedback control is used, then control speed is improved, but positioning accuracy deteriorates due to substrate variations
Solution Approach 1:
The system implements dual feedback mechanisms: encoder-based feedback for high-speed control responses and image-based feedback for high-precision positioning verification. The switching unit determines which feedback source to prioritize based on the operational phase, allowing the system to benefit from both fast encoder responses and accurate image-based positioning data.
Data Source
AI summary
A motor control device includes a feedforward controller generating motor torque and position commands, an encoder outputting a motor position detection value, a machine-end sensor detecting a target object as a measurement value, a signal processor calculating movable component position based on the measurement value as a machine-end position detection value, and a feedback torque command generation unit generating a feedback torque command based on the machine-end and motor position detection values and the motor position command. When a positional relationship between the movable component and a target point is undetectable, the feedback torque command is generated based on the motor position detection value and the motor position command, and when the positional relationship is detectable, the feedback torque command is generated based on the motor and machine-end position detection values, the motor position command, and a signal adding a time delay to the motor position command.