Constraint Learning From Success-Failure Time Series in Control

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing control systems for robots and other control targets require manual setting of constraint conditions, which can be burdensome and may not account for unforeseen constraints.

Innovation Solution

A constraint condition acquisition device and method that acquires time-series data from successful and failed tasks, uses a template acquisition mechanism to identify constraint condition templates, and calculates parameter values to define constraint conditions that hold in successful data and do not hold in failed data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If constraint conditions are manually set in advance for control, then control reliability is improved, but operator burden increases and unforeseen constraints cannot be accounted for

Engineering Contradiction:

Improvecontrol reliabilityVSAvoidoperator burden

Core Design Contradiction:

ReliabilityVSEase of operation

Solution Approach 1:

The control device automatically acquires constraint conditions by analyzing time-series data from successful and failed tasks, eliminating the need for manual operator input. The system self-learns constraints through data processing, thereby reducing operator burden while maintaining control reliability through automated constraint acquisition

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system uses feedback from task execution results (success/failure time-series data) to automatically update and refine constraint conditions. By continuously learning from operational outcomes, the system improves control reliability while requiring minimal operator intervention beyond initial data collection

Inventive Principle:

Principle #23Feedback

2Ease of operation

If preset constraint conditions are used, then control can be performed with simple settings, but constraints that have not been clarified cannot be handled

Engineering Contradiction:

Improvecontrol simplicityVSAvoidconstraint coverage

Core Design Contradiction:

Ease of operationVSAdaptability or versatility

Solution Approach 1:

The constraint conditions transition from static preset values to dynamic, data-driven parameters. The system continuously updates constraint conditions based on analyzed time-series data from task executions, enabling the control system to adapt to newly discovered constraints while maintaining simple operation through automated processes

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system performs preliminary analysis of time-series data from successful and failed tasks to pre-identify potential constraint conditions before actual control execution. This preliminary constraint acquisition enables the system to handle previously unclarified constraints while maintaining operational simplicity

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20250130562A1Constraint condition acquisition device, control system, constraint condition acquisition method, and recording medium

Publication Date: 2025.04.24 NEC CORP

AI summary

A constraint condition acquisition device acquires success time time-series data, which is time-series data pertaining to control over a control target when a predetermined task carried out through the control is successful, and failure time time-series data, which is the time-series data in the case of task failure; acquires a constraint condition template, which is a constraint condition that includes a parameter; and determines the value of the parameter such that the constraint condition holds in the success time-series data and the constraint condition does not hold in the failure time-series data.