Emotion-Aware Robot Action Control for User Behavior Response

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing action control systems for robots fail to appropriately respond to user emotions and behaviors, leading to potential discomfort or misinterpretation.

Innovation Solution

An action control system that includes a user state recognition unit, emotion determination unit, and action determination unit to assess user emotions and behaviors, using reaction rules and thresholds to determine appropriate actions, including imitating positive behaviors and addressing negative emotions through defined actions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the robot executes actions based on emotion values for positive emotions, then interaction quality improves, but complexity of determining appropriate actions increases

Engineering Contradiction:
Improveinteraction qualityVSAvoidaction determination complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system changes the parameter of emotion value (positive/negative classification) to determine action type. When emotion value is positive, the robot executes actions corresponding to user behavior; when negative, it executes actions based on reaction rules. This parameter-based differentiation simplifies the decision-making process while maintaining high adaptability.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The action determination process is segmented into two distinct pathways based on emotion value: one for positive emotions (behavior-corresponding actions) and one for negative emotions (reaction rule-based actions). This segmentation reduces overall complexity by dividing the decision space into manageable segments.

Inventive Principle:
Principle #1Segmentation

2Reliability

If the robot uses reaction rules for negative emotions, then appropriate responses are provided, but loss of information about user intent increases

Engineering Contradiction:
Improveresponse appropriatenessVSAvoiduser intent information
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

Reaction rules serve as an intermediary mechanism that bridges the gap between negative emotion detection and appropriate response generation. The rules contain pre-defined mappings from behavior classifications to suitable actions, preserving user intent information that would otherwise be lost in raw negative emotion data.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

Reaction rules are prepared in advance with predefined actions for various negative emotion scenarios. This preliminary preparation ensures that when negative emotions are detected, the robot can quickly retrieve and execute appropriate responses without losing critical user intent information.

Inventive Principle:
Principle #10Preliminary action

3Productivity

If the robot imitates positive user behaviors, then user engagement increases, but energy consumption increases

Engineering Contradiction:
Improveuser engagementVSAvoidrobot energy consumption
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The robot performs partial imitation of user behaviors by executing actions that correspond to the classified behavior type rather than fully replicating the physical behavior. For example, if the user smiles, the robot may activate a smile expression and select a friendly response action, rather than physically mimicking the facial movement, thus reducing energy consumption while maintaining engagement.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP4682759A1Action control system and program
Publication Date: 2026.01.21 SOFTBANK GROUP CORP
  • EP4682759A1 patent drawingFigure 1
  • EP4682759A1 patent drawingFigure 2
  • EP4682759A1 patent drawingFigure 3

AI summary

An action control system includes: a user state recognition unit that recognizes a user state including a behavior of a user; an emotion determination unit that determines an emotion value indicating an emotion of the user on the basis of the user state; and an action determination unit that determines an action corresponding to the behavior of the user as an action to be executed in a case where the emotion value is a positive value, and determines an action determined on the basis of a reaction rule in accordance with a behavior classification of the user as an action to be executed in a case where the emotion value is not a positive value.