Driver State Estimation with Dual Attention Indices

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing driver state determination devices struggle to accurately estimate a driver's attention function degraded state, which is crucial for safe driving, especially in cases where the driver's focus is on a single point, and fail to account for conditions caused by diseases or health issues.

Innovation Solution

A driver state estimation device that utilizes both top-down and bottom-up attention source amounts by detecting a driver's sightline and deriving index values based on motion relative to points of interest and saliency distribution, allowing for improved estimation accuracy by comparing these values against set thresholds.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If driver state is estimated based on sightline concentration on one point and distance to preceding vehicle, then absent-minded state can be detected, but attention function degraded state caused by disease cannot be estimated

Engineering Contradiction:
Improvedriver state estimation accuracyVSAvoidability to detect different types of attention states
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent segments attention into two distinct types: top-down attention (goal-directed, voluntary) and bottom-up attention (stimulus-driven, involuntary). By measuring both types separately using different指标 (first index for top-down, second index for bottom-up), the system can distinguish between absent-minded states (affecting top-down) and disease-caused attention degradation (affecting bottom-up), thereby resolving the contradiction between detection precision and adaptability.

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If only top-down attention source amount is used for estimation, then goal-directed attention can be measured, but stimulus-driven attention response cannot be captured

Engineering Contradiction:
Improvetop-down attention measurement accuracyVSAvoidbottom-up attention information
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The patent merges two complementary measurement approaches: top-down attention measurement (based on sightline concentration and task relevance) and bottom-up attention measurement (based on salient stimulus detection). By combining both indices, the system captures the complete attention picture, preventing information loss while maintaining precise measurement of goal-directed attention.

Inventive Principle:
Principle #5Merging (Combining)

3Device complexity

If driver state determination relies on single指标, then system complexity is reduced, but estimation accuracy deteriorates

Engineering Contradiction:
Improvedetermination system structureVSAvoiddriver state estimation accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent adds a new dimension to driver state determination by introducing dual-index measurement (first index for top-down attention, second index for bottom-up attention) instead of relying on a single指标. This dimensional expansion enables more accurate estimation of different attention states without creating excessive system complexity, as each index serves a distinct functional purpose.

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

Data Source

PatentEP3868621B1Driver state estimation device and movable body
Publication Date: 2023.10.18 MAZDA MOTOR CORP
  • EP3868621B1 patent drawingFigure 1
  • EP3868621B1 patent drawingFigure 2
  • EP3868621B1 patent drawingFigure 3

AI summary

A first measurement section measures a first index value that is correlated with an amount of an attention source allocated to top-down attention of a driver. A second measurement section measures a second index value that is correlated with an amount of the attention source allocated to bottom-up attention of the driver. An estimation section estimates the driver state including an attention function degraded state of the driver on the basis of the first index value and the second index value.