Head-wearable Device for Uniform Motion-based Behavior Estimation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing mobile device technologies limit the types of user behavior that can be estimated due to non-uniform holding or usage methods, requiring extensive sensor data for accurate estimation.

Innovation Solution

A head-wearable device, such as a glasses-type HMD, equipped with gravity acceleration sensors, gyro sensors, and an imaging lens, estimates user behavior based on head motion, transmitting data to a server for characteristic analysis and feedback control.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If mobile phone terminals or smartphones are used for behavior recognition, then existing devices can be utilized without additional equipment, but the types of user behavior that can be estimated are limited and estimation accuracy is reduced due to non-uniform holding or usage methods

Engineering Contradiction:
Improvetypes of user behavior that can be estimatedVSAvoidestimation accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent introduces a head-wearable device as an intermediary between the user and the behavior recognition system. This device is positioned on the head, which is a body part that moves uniformly during various activities. The head-wearable device captures motion data that serves as a reliable mediator for estimating user behavior, overcoming the limitations of handheld devices whose positions and orientations vary significantly during use.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent changes the reference frame for motion detection from handheld devices to head-mounted devices. By mounting sensors on the head, the system captures motion parameters (acceleration, angular velocity) that directly reflect body movement patterns. This parameter change enables more accurate extraction of behavior characteristics because head motion more uniformly represents overall body movement compared to handheld device motion.

Inventive Principle:
Principle #35Parameter changes

2Device complexity

If mobile phone terminals are used for behavior recognition, then device complexity is minimized by using existing equipment, but a large number of sensor data pieces are required to improve estimation accuracy

Engineering Contradiction:
Improvedevice configurationVSAvoidnumber of sensor data pieces required
Core Design Contradiction:
Device complexityVSQuantity of substance

Solution Approach 1:

The head-wearable device serves as an effective intermediary that captures motion data with higher behavioral relevance. Because the head moves more uniformly during activities compared to handheld devices, the motion data collected requires fewer data points to achieve accurate behavior estimation. This reduces the quantity of sensor data needed while maintaining or improving estimation accuracy.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

By changing from handheld to head-mounted sensing, the system captures motion parameters that are more directly correlated with behavior patterns. This parameter change increases the information content per data point, reducing the total number of sensor readings required to achieve reliable behavior recognition.

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If head-wearable devices are used for behavior recognition, then estimation accuracy is improved through uniform head motion detection, but device complexity increases compared to mobile phones

Engineering Contradiction:
Improvebehavior estimation accuracyVSAvoiddevice structure
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts the essential behavior recognition function from complex mobile phone systems and implements it in a simplified head-wearable device. By taking out only the necessary sensors (accelerometer, gyro sensor) and processing capabilities needed for motion-based behavior recognition, the system achieves accurate estimation with reduced overall device complexity compared to full-featured smartphones.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent replaces the complex interaction mechanics of handheld devices (which require users to hold and manipulate the device in various ways) with a simple head-wearing mechanism. This substitution eliminates the variability introduced by different holding methods while maintaining the essential sensing function, achieving better accuracy with simpler device-user interaction.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Robustly estimates behaviors like talking, eating, or napping, providing feedback and safety measures, and enhancing communication by uniformly detecting head motion across users, reducing the need for varied sensor data and improving estimation accuracy.

Implementation Method 1

using a gravity acceleration sensor that detects motion of the head

Methodology Applied
Scientific EffectGravity: Gravitation

Implementation Method 2

a gyro sensor which detects motion of the head

Methodology Applied
Scientific EffectGyroscopic effect: Gyroscope

Data Source

PatentEP3067782B1Information processing apparatus, control method, and program
Publication Date: 2021.05.05 SONY GROUP CORP
  • EP3067782B1 patent drawingFigure 1
  • EP3067782B1 patent drawingFigure 2
  • EP3067782B1 patent drawingFigure 3

AI summary

[Object] To provide an information processing apparatus, a control method and a program which enable estimation of user behavior based on motion of the head. [Solution] An information processing apparatus includes an estimating unit configured to estimate behavior of a user by comparing data detected by a detecting unit worn on a head of the user and configured to detect motion of the head with a behavior characteristic amount calculated based on learning data, and a control unit configured to perform control according to the behavior estimated by the estimating unit.