Controller Parameter Updating for Sectional and Instantaneous Errors

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing controller parameter updating methods fail to effectively address control errors caused by temporary factors, such as frictional forces during the start of movement, as they evaluate errors based on entire time-series data rather than specific conditions.

Innovation Solution

A parameter updating method that calculates and updates variable parameters using an evaluation function comprising parts for entire, sectional, and instantaneous errors, allowing for targeted reduction of control errors, particularly in low velocity sections where frictional forces significantly impact accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If controller parameters are updated based on entire time-series data, then overall control accuracy is improved, but control errors caused by temporary causes (e.g., friction at start) cannot be appropriately addressed

Engineering Contradiction:
Improvecontrol error evaluation accuracyVSAvoidcontroller parameter update reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The observation period is divided into multiple sections based on control output values or time intervals. The evaluation function calculates sectional errors for each divided section, allowing temporary errors in specific sections (e.g., friction at start) to be identified and handled separately from overall control performance, thus preventing them from unduly influencing parameter updates.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The evaluation function incorporates multiple error components (entire error, sectional errors, and instantaneous errors) with different weights. By adjusting weights, the system can emphasize certain error types (e.g., sectional errors for temporary causes) over others, enabling selective focus on specific control error sources without being overwhelmed by all errors equally.

Inventive Principle:
Principle #16Partial or excessive action

2Device complexity

If the evaluation function considers only entire error, then parameter update is simple, but temporary control errors are not distinguished from systematic errors

Engineering Contradiction:
Improveevaluation function complexityVSAvoiderror source discrimination capability
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The observation period is divided into multiple sections based on control output values or time intervals. The evaluation function calculates sectional errors for each divided section, allowing temporary errors in specific sections (e.g., friction at start) to be identified and handled separately from overall control performance, thus preventing them from unduly influencing parameter updates.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The evaluation function incorporates multiple error components (entire error, sectional errors, and instantaneous errors) with different weights. By adjusting weights, the system can emphasize certain error types (e.g., sectional errors for temporary causes) over others, enabling selective focus on specific control error sources without being overwhelmed by all errors equally.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS11449031B2Parameter updating method, parameter updating system, and storage medium storing program
Publication Date: 2022.09.20 BROTHER KOGYO KK
  • US11449031B2 patent drawing
  • US11449031B2 patent drawing
  • US11449031B2 patent drawing

AI summary

A method includes: acquiring time-series data of a control input and a control output; calculating a variable parameter that minimizes an evaluation function, based on observation values of the control input and the control output; and updating the variable parameter with the calculated variable parameter. The evaluation function includes: a first function part that evaluates an entire error that is a control error of the entire observation period and increases the output value as the entire error increases; and at least one of: a second function part that evaluates a sectional error that is a control error of a particular section in the observation period and increases the output value as the sectional error increases; and a third function part that evaluates an instantaneous error that is a control error of a particular time point in the observation period and increases the output value as the instantaneous error increases.