Automated Driver Gaze Analysis for Workload Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for analyzing driver eye movements are labor-intensive, lack standardization, and fail to account for smooth pursuits, making it difficult to assess driver distraction and workload effectively, which contributes to vehicular collisions.

Innovation Solution

A method for automated analysis of behavioral movement data using validated algorithms that classify gaze-direction instances as on-location or off-location, allowing for real-time detection of elevated driver workload and distraction, and providing standardized measures in accordance with ISO/SAE standards, utilizing sensors like cameras and ultrasonic devices to track head and eye movements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual analysis methods are used to study driver eye movements, then detailed behavioral data can be collected, but the analysis process becomes labor-intensive and time-consuming

Engineering Contradiction:
Improveeye movement analysis accuracyVSAvoidanalysis time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces manual mechanical analysis of eye movement video data with automated computational algorithms. The system uses computer-based processing to automatically detect, classify, and analyze fixation and saccade events from eye tracking data, eliminating the need for manual frame-by-frame review while maintaining measurement precision.

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

Solution Approach 2:

The analysis system performs self-service by automatically processing eye movement data without requiring external manual intervention. The computational algorithms independently identify visual fixations, calculate fixation durations, and generate analysis results, making the system self-sufficient and eliminating labor-intensive manual operations.

Inventive Principle:
Principle #25Self-service

2Reliability

If comprehensive eye movement tracking is implemented, then driver cognitive state can be assessed, but device complexity and cost increase

Engineering Contradiction:
Improvedriver state assessment accuracyVSAvoidtracking system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts and focuses on specific, critical eye movement parameters (fixation locations, durations, and patterns) rather than attempting to analyze all eye movement aspects. By isolating the most relevant features for driver state assessment, the system reduces computational complexity while maintaining reliability in detecting driver distraction and cognitive load.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The eye tracking system serves multiple functions: detecting driver distraction, assessing cognitive load, monitoring alertness, and evaluating visual attention distribution. This multi-functionality justifies the device complexity by providing comprehensive driver state assessment through a single integrated system rather than multiple separate devices.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Speed

If real-time eye movement analysis is performed, then driver distraction can be detected immediately, but processing requirements and computational load increase

Engineering Contradiction:
Improvedistraction detection speedVSAvoidcomputational energy consumption
Core Design Contradiction:
SpeedVSUse of energy by moving object

Solution Approach 1:

The system performs preliminary classification of eye movement data into fixations and saccades using simple, pre-defined criteria before conducting more complex analysis. This preliminary sorting reduces the computational burden of subsequent real-time processing by organizing data into manageable categories that require less intensive computation.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements real-time analysis of only the most critical eye movement parameters necessary for distraction detection, rather than performing complete comprehensive analysis on all eye movement features. This partial analysis approach maintains sufficient detection speed and reliability while reducing computational energy consumption to acceptable levels.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP2032034B1Method for determining and analyzing a location of visual interest
Publication Date: 2020.04.01 VOLVO TRUCK CORP
  • EP2032034B1 patent drawingFigure 1
  • EP2032034B1 patent drawingFigure 2~3
  • EP2032034B1 patent drawingFigure 4~5

AI summary

A method on analyzing data based on the physiological orientation of a driver is provided. Data is descriptive of a driver's gaze-direction is processing and criteria defining a location of driver interest is determined. Based on the determined criteria, gaze-direction instances are classified as either on-location or off-location. The classified instances can then be used for further analysis, generally relating to times of elevated driver workload and not driver drowsiness. The classified instances are transformed into one of two binary values (e.g., 1 and 0) representative of whether the respective classified instance is on or off location. The uses of a binary value makes processing and analysis of the data faster and more efficient. Furthermore, classification of at least some of the off-location gaze direction instances can be inferred from the failure to meet the determined criteria for being classified as an on-location driver gaze direction instance.