Driver Sightline Analysis for Distinguishing Distracted States

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing driver state determination systems fail to accurately distinguish between a distracted state and other abnormal states, such as diseases, due to reduced changes in face direction, sightline direction, or visual recognition behavior, leading to inaccurate determinations.

Innovation Solution

A driver state determination apparatus that utilizes a travel environment information acquisition device, sightline detector, and controller to assess driving load scores, distracted state occurrence scores, search behavior scores, and cognitive/operation load scores, distinguishing distracted states from abnormal states by analyzing visual perception and environmental factors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional techniques focus on change amounts of face direction and sightline direction or number of times of visual recognition behavior, then detection simplicity is maintained, but measurement precision deteriorates because it is impossible to distinguish distracted state from other abnormal states like disease

Engineering Contradiction:
Improveaccuracy of distracted state determinationVSAvoidcomplexity of determination system
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The determination process is segmented into multiple independent scoring stages: driving load score acquisition, distracted state occurrence score acquisition, search behavior score acquisition, and distracted state level acquisition. Each stage processes specific parameters independently, allowing comprehensive analysis without requiring a single complex determination system.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system transitions from two-dimensional detection (face direction and sightline direction changes) to multi-dimensional analysis by incorporating driving load, distracted state occurrence probability, and search behavior normality. This dimensional expansion enables distinction between distracted states and other abnormal states through patterns across multiple parameters.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If the system uses multiple scoring parameters and comprehensive analysis, then measurement precision improves for distinguishing distracted states, but device complexity increases

Engineering Contradiction:
Improveaccuracy of distracted state determinationVSAvoidcomplexity of determination system
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The determination process is segmented into multiple independent scoring stages: driving load score acquisition, distracted state occurrence score acquisition, search behavior score acquisition, and distracted state level acquisition. Each stage processes specific parameters independently, allowing comprehensive analysis without requiring a single complex determination system.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system transitions from two-dimensional detection (face direction and sightline direction changes) to multi-dimensional analysis by incorporating driving load, distracted state occurrence probability, and search behavior normality. This dimensional expansion enables distinction between distracted states and other abnormal states through patterns across multiple parameters.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Productivity

If the system promptly determines distracted state based on driving load and occurrence probability, then productivity improves, but measurement precision may worsen due to potential false positives

Engineering Contradiction:
Improvespeed of distracted state determinationVSAvoidaccuracy of distracted state determination
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

The system performs preliminary assessment by calculating driving load score and distracted state occurrence probability first. When the occurrence probability exceeds a threshold, it triggers more detailed search behavior analysis. This preliminary action enables prompt identification of high-risk situations while maintaining precision through subsequent verification.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system uses feedback loops where search behavior scores are continuously updated based on comparison between predicted and actual sightline directions. This feedback mechanism allows the system to confirm or refute initial distracted state predictions, maintaining high precision while enabling prompt determination through iterative refinement.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4342758B1Driver state determination apparatus
Publication Date: 2026.03.11 MAZDA MOTOR CORP
  • EP4342758B1 patent drawingFigure 1
  • EP4342758B1 patent drawingFigure 2
  • EP4342758B1 patent drawingFigure 3

AI summary

[Task] To provide a driver state determination apparatus capable of distinguishing a driver's distracted state from another abnormal state such as a disease and promptly and accurately determining the driver's distracted state. [Solution] A driver state determination apparatus 100 includes a controller 10 configured to determine a driver's state on the basis of travel environment information of a vehicle and the driver's sightline. The controller 10 acquires a driving load score on the basis of the travel environment information, acquires a distracted state occurrence score, which represents a degree of likelihood that the driver is brought into the distracted state, on the basis of the driving load score and an elapsed time with the driving load score, acquires a search behavior score, which represents a degree of normality of search behavior by the driver's visual perception, on the basis of the travel environment information and the driver's sightline when the distracted state occurrence score is equal to or higher than a specified value, acquires a distracted state level of the driver on the basis of the search behavior score and an elapsed time with the search behavior score, and determines that the driver is in the distracted state when the distracted state level is equal to or higher than a threshold and the search behavior score is increased in response to an increase in the driving load score.