Drowsy Driving Detection via Eye Movement and Steering Torque Analysis

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current drowsy driving warning systems lack accuracy in detecting drowsiness at the onset and fail to provide timely warnings, with indirect methods being less accurate and direct methods risking false alarms due to long eye closure detection.

Innovation Solution

A drowsy driving management device that utilizes a processor to detect slow eye movement and lack of steering torque changes, calculating eye movement information and determining drowsiness based on fixation and head pose data, with a warning system providing visual, tactile, or audible alerts when drowsiness is confirmed.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If indirect manner based on driving pattern is used to determine drowsy driving, then device complexity is reduced, but measurement precision of drowsy driving detection deteriorates

Engineering Contradiction:
Improvedetection system complexityVSAvoiddrowsy driving detection accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent segments the drowsy driving detection into multiple independent analysis modules: eye movement analysis (calculating eye movement distance from gaze and head pose data), steering torque analysis (detecting intervals with no change in steering torque), and user maneuver analysis. Each module processes specific data independently, then their results are combined to determine drowsy driving, reducing overall system complexity while maintaining high detection accuracy through multi-faceted analysis.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If direct manner based on eye opening/closing image is used to determine drowsy driving, then measurement precision of drowsy driving detection is improved, but false alarm rate increases due to long eye closure detection

Engineering Contradiction:
Improvedrowsy driving detection accuracyVSAvoidwarning timing accuracy
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent performs preliminary analysis of eye movement patterns before issuing a drowsy driving warning. By calculating eye movement distance from gaze and head pose data over time, the system identifies slow eye movement as an early indicator of drowsiness before complete eye closure occurs. This preliminary detection action allows the system to warn drivers at the onset of drowsiness rather than waiting for definitive eye closure, improving warning timing accuracy and reducing false alarms.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces steering torque analysis as an intermediary indicator to confirm drowsy driving state. Instead of relying solely on eye closure detection, the system uses steering torque changes as a mediating parameter - when both slow eye movement and lack of steering torque variation occur simultaneously, the confidence in drowsy driving determination increases. This intermediary check reduces false alarms while maintaining early detection capability.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Measurement precision

If warning is issued based on long eye closure detection, then measurement precision of drowsy driving detection is improved, but loss of time for early warning increases

Engineering Contradiction:
Improvedrowsy driving detection accuracyVSAvoidwarning delay
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary analysis of eye movement patterns before issuing a drowsy driving warning. By calculating eye movement distance from gaze and head pose data over time, the system identifies slow eye movement as an early indicator of drowsiness before complete eye closure occurs. This preliminary detection action allows the system to warn drivers at the onset of drowsiness rather than waiting for definitive eye closure, improving warning timing accuracy and reducing false alarms.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent uses partial eye movement information (eye movement distance calculated from gaze and head pose) rather than requiring complete eye closure detection. By analyzing partial indicators of drowsiness (slow eye movement combined with steering torque patterns) rather than waiting for the full manifestation of eye closure, the system achieves earlier detection with acceptable accuracy, reducing warning delay while maintaining reliability.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS10882533B2Drowsy driving management device, system including the same, and method thereof
Publication Date: 2021.01.05 HYUNDAI MOTOR CO LTD
  • US10882533B2 patent drawing
  • US10882533B2 patent drawing
  • US10882533B2 patent drawing

AI summary

A drowsy driving management device includes: a processor configured to determine whether slow eye movement of a user occurs and whether there is no change in steering torque and to determine that the user drives while drowsy when the slow eye movement of the user occurs and when there is no change in the steering torque for a predetermined period of time; and a storage that stores information indicating whether the slow eye movement occurs and a result of determining whether there is no change in the steering torque.