Activity Recognition via Static Electric Field and Accelerometer Fusion

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional methods for detecting user activities using electronic devices are limited by the need for significant motion of the device, making it difficult to accurately track activities like bicycling, where arm motion is minimal.

Innovation Solution

An electronic device employing an accelerometer in combination with a static electric field sensor to form a characteristic signature by comparing data from both sensors with predefined baseline signatures, allowing for accurate identification of user activities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If an accelerometer is used to detect user activity, then motion-based activities can be detected, but activities with minimal arm motion (like bicycling) cannot be accurately detected

Engineering Contradiction:
Improveactivity detection accuracyVSAvoiddetection capability across different activity types
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent combines data from a static electric field sensor and an accelerometer to form a composite characteristic signature. The static electric field sensor detects changes in the electric field caused by body movement, while the accelerometer detects motion. By merging these two data sources, the system can accurately detect both high-motion activities (running, walking) and low-motion activities (bicycling, swimming) where the electric field changes provide complementary information to the accelerometer data.

Inventive Principle:
Principle #5Merging (Combining)

2Ease of operation

If the electronic device is worn on the wrist to detect activity, then arm motion can be detected, but activities where the arm is stationary (like bicycling) become difficult to detect

Engineering Contradiction:
Improvedevice wearabilityVSAvoidactivity detection accuracy for stationary arm activities
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The static electric field sensor acts as an intermediary that detects body movement indirectly through changes in the electric field surrounding the body, rather than requiring direct mechanical motion of the device. This allows the system to detect activities like bicycling where the arm and wrist remain relatively stationary, as the electric field changes capture subtle body movements that occur during such activities.

Inventive Principle:
Principle #24Intermediary (Mediator)

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

Enhances the detection of physical activities by accurately identifying user actions, including stationary or low-motion activities, through the combination of static electric field and motion data, improving the device's ability to determine the type of activity being performed.

Implementation Method 1

a static electric field sensor for detecting a static electric field

Methodology Applied
Scientific EffectStatic electric field sensing: Electric Field

Implementation Method 2

at least one other sensor for detecting one of motion or sound

Methodology Applied
Scientific EffectAccelerometer sensing: Accelerometer

Data Source

PatentUS10605595B2Biking activity recognition through combination of electric field sensing and accelerometer
Publication Date: 2020.03.31 SONY GROUP CORP
  • US10605595B2 patent drawing
  • US10605595B2 patent drawing
  • US10605595B2 patent drawing

AI summary

An apparatus and method for determining a user activity include or define a plurality of baseline signatures, each baseline signature corresponding to a type of user activity and having data formed from a first data representing a varying static electric field and a second data representing motion. Data responsive to a varying static electric field is obtained from a first sensor, and data responsive to motion is obtained from a second sensor. The first data is combined with the second data, and the user activity is identified based on a comparison of the combined first and second data with the plurality of baseline signatures.