Robot Action Control Using Emotion-Aware Conversation Responses

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing technologies lack the ability for robots to execute appropriate actions in response to user questions, particularly in determining and expressing emotions accurately.

Innovation Solution

An action control system that includes an answer generation unit, emotion determination unit, and action determination unit to generate answers, determine user emotions, and execute specific actions based on those emotions, such as gestures and movements, using a combination of chat and emotion engines.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a robot executes actions based on basic conversation functions, then the robot can respond to user questions, but the robot cannot accurately determine and express appropriate emotions in response

Engineering Contradiction:
Improveemotion determination accuracyVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system is divided into distinct functional modules: a conversation function for text processing, an emotion estimation function for emotional state detection, and an action determination function for selecting appropriate responses. Each module handles a specific aspect of the interaction, allowing the system to accurately determine emotions without requiring a complete redesign of the entire system architecture.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The emotion estimation function serves as an intermediary between the conversation function and the action determination function. It processes the text from the conversation function and transforms it into emotional state information that the action determination function can use to select appropriate robot actions, enabling accurate emotion determination while maintaining modular system design.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If the robot uses multiple functions (conversation, emotion estimation, action determination), then the robot can execute appropriate actions, but the system becomes more complex

Engineering Contradiction:
Improveresponse adaptabilityVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The robot system integrates multiple functions including conversation processing, emotion estimation, and action determination within a single unified platform. This multi-functional approach allows the robot to adapt its responses based on both the content and emotional tone of user inputs, enhancing response adaptability while consolidating functions into one system rather than requiring separate devices for each function.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The conversation function, emotion estimation function, and action determination function are merged into an integrated system where the output of one function serves as input to the next. The text processing output feeds into emotion estimation, which in turn feeds into action determination, creating a streamlined workflow that enhances versatility without requiring separate independent systems for each function.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP4684927A1Action control system, information presentation system, program, and robot
Publication Date: 2026.01.28 SOFTBANK GROUP CORP
  • EP4684927A1 patent drawingFigure 1
  • EP4684927A1 patent drawingFigure 2
  • EP4684927A1 patent drawingFigure 3

AI summary

An action control system includes: an answer generation unit that generates an answer to a question of a user based on a conversation function of causing a robot and the user to have a conversation with each other; an emotion determination unit that determines an emotion of the user in response to the answer based on an emotion estimation function of estimating the emotion of the user; and an action determination unit that determines an action of the robot corresponding to the emotion.