Robot Emotion Recognition for Over-Immersion Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Communication robots fail to effectively detect and mitigate over-immersion states in users, particularly children, which can lead to negative emotional and health impacts due to prolonged engagement with content.
Innovation Solution
A robot equipped with an output interface, a camera, and a processor that uses an emotion recognizer trained through deep learning to detect over-immersion by analyzing user images, tracking gaze, and measuring eye blinks, and takes corrective actions such as displaying touch items, rotating, or tilting to release the user from over-immersion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the robot continuously outputs content to engage the user, then user engagement and interaction are improved, but the user may enter an over-immersion state causing negative emotional and health impacts
Solution Approach 1:
The robot employs a feedback mechanism by continuously monitoring the user's emotional state through the emotion recognizer during content output. When the recognizer detects over-immersion emotions (such as excessive excitement or stress), the robot adjusts or stops content output to prevent harmful effects, thus resolving the contradiction between maintaining engagement and preventing over-immersion harm
Solution Approach 2:
The robot implements periodic emotional state detection during content output, rather than continuous monitoring. The emotion recognizer is activated at regular intervals to assess user engagement levels, allowing the robot to maintain content output efficiency while periodically checking for and preventing over-immersion states
2Measurement precision
If the robot uses deep learning emotion recognition to accurately detect user emotions, then detection precision is improved, but device complexity increases
Solution Approach 1:
The patent introduces an emotion recognizer as an intermediary component that specializes in emotional analysis. This dedicated module processes facial expressions and physiological signals to detect emotions, separating the complex detection function from the main robot control system. The emotion recognizer uses deep learning algorithms internally while presenting a simplified interface to the robot's control logic
Solution Approach 2:
The robot system is segmented into distinct functional modules: content output interface, emotion recognizer, and control unit. The emotion detection function is isolated in the emotion recognizer module, which handles the complex deep learning computations separately from the main control system. This segmentation allows high detection precision while managing device complexity through modular architecture
3Reliability
If the robot monitors user state continuously to detect over-immersion, then detection reliability is improved, but energy consumption increases
Solution Approach 1:
The robot implements periodic monitoring of the user's emotional state during content output, rather than continuous monitoring. The emotion recognizer is activated at regular intervals to assess engagement levels, which maintains detection reliability by catching over-immersion states while significantly reducing energy consumption compared to continuous operation
Solution Approach 2:
The monitoring frequency is dynamically adjusted based on the current content output state and detected emotional patterns. When the user shows signs of increasing engagement, the robot increases monitoring frequency to detect over-immersion earlier. When engagement is stable and low, monitoring frequency is reduced to conserve energy, thus balancing reliability and energy consumption
Data Source
AI summary
Disclosed herein is a robot including an output interface including at least one of a display or a speaker, a camera, and a processor controlling the output interface to output content, acquiring an image including a user through the camera while the content is output, detecting an over-immersion state of the user based on the acquired image, and controlling an operation of releasing over-immersion when the over-immersion state is detected.