Cost Function Adjustment for Post-Change Device Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems struggle to efficiently adjust parameters in response to changes in device environments, such as camera position or sensor type, due to unknown sensor characteristics and environmental dependencies, leading to time-consuming recalibration.
Innovation Solution
An adjustment system that utilizes an input unit for pre-change and post-change performance data, and an update unit to generate a second cost function by estimating errors in the output values of the first cost function, allowing for efficient parameter adjustment without requiring full recalibration of devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If calibration is performed in advance to associate machine coordinate system with camera coordinate system, then control accuracy is improved, but when environment changes (camera position or new camera), the system cannot control devices appropriately and requires time-consuming parameter verification
Solution Approach 1:
The patent transforms the static calibration parameters into dynamic adjustable parameters. The learning model continuously learns and adapts to environmental changes by processing demonstration data, allowing the system to maintain control accuracy without manual recalibration when camera position or equipment changes.
Solution Approach 2:
The patent changes the parameters of the learning model based on demonstrated operations rather than fixed calibration parameters. By learning from demonstration data, the system automatically adjusts parameters to adapt to environmental changes, eliminating the need for time-consuming parameter verification.
2Adaptability or versatility
If parameters of sensors and control models are modified in response to environmental changes, then adaptability is improved, but since sensor characteristics are often unknown, device parameter verification is required which leads to time-consuming adjustment
Solution Approach 1:
The learning model performs self-adjustment by automatically learning from demonstration data. When environmental changes occur, the system autonomously updates its parameters through continuous learning without requiring manual verification or adjustment of sensor characteristics, significantly reducing parameter adjustment time.
Solution Approach 2:
The system uses feedback from demonstrated operations to continuously improve its control parameters. By processing demonstration data and comparing actual performance with expected outcomes, the learning model automatically adjusts parameters to maintain optimal performance without time-consuming manual verification.
3Measurement precision
If device characteristics are verified by itself when making changes, then measurement precision is maintained, but the adjustment process requires a lot of time due to trade-off considerations among various factors
Solution Approach 1:
The system performs preliminary learning by continuously processing demonstration data in advance. This preliminary action builds a robust learning model that can quickly adapt to changes without requiring time-consuming verification, as the model has already learned from extensive demonstration data covering various operational conditions.
Data Source
AI summary
An input unit 81 receives inputs of pre-change performance data acquired by a device before a change and post-change performance data acquired by the device after having undergone the change, through control using a first cost function. An update unit 82 generates a second cost function obtained by updating the first cost function in such a way as to reduce a difference between the pre-change performance data and the post-change performance data. In the process, the update unit 82 generates the second cost function obtained by updating the first cost function by estimating an error that occurs in an output value of the device included in the first cost function before and after the change to the device.