Driver Monitoring Gaze Prediction Using Virtual Focus Points

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current Driver Monitoring Systems (DMS) fail to effectively address unusual driver conditions such as visual impairments and overlook the relationship between driver gaze and the external environment, leading to inaccurate attention assessment in complex driving scenarios.

Innovation Solution

A dual-sensor system integrating interior and exterior sensors with a computing unit that analyzes data from both to derive a driver's gaze vector by assigning a virtual focus point based on perceived changes, using autocalibration and machine learning to refine gaze predictions, and adapt to individual driver conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If a single interior camera is used to monitor driver behavior, then the system cost is reduced and device complexity is lowered, but measurement precision of driver attention and gaze accuracy deteriorates

Engineering Contradiction:
Improvesystem complexityVSAvoidgaze vector accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent combines interior camera data with exterior camera data to estimate the driver's gaze vector. By merging information from both interior and exterior sensors, the system achieves more accurate gaze prediction than would be possible with a single interior camera alone, while avoiding the complexity of multiple interior cameras.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces virtual focus points as an intermediary element to bridge interior and exterior sensor data. These virtual focus points represent locations in the external environment that the driver is likely looking at, based on head pose and environmental features, enabling accurate gaze estimation without direct eye tracking.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Device complexity

If traditional DMS focuses only on interior monitoring, then device complexity is reduced, but adaptability to diverse driver conditions and external interactions deteriorates

Engineering Contradiction:
Improvesensor configurationVSAvoiddriver condition coverage
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent makes the DMS system universal by integrating both interior and exterior monitoring capabilities. The system can handle diverse driver conditions (drowsiness, distraction, visual impairments) and external interactions (road signs, traffic signals, pedestrians) through a unified approach that combines data from multiple sensor types, enhancing overall adaptability.

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

3Measurement precision

If DMS overlooks the relationship between driver gaze and external environment, then measurement precision of internal states is maintained, but reliability of attention assessment in complex driving scenarios deteriorates

Engineering Contradiction:
Improvegaze detection accuracyVSAvoidattention assessment reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent implements feedback by continuously comparing the estimated gaze vector with actual driver behavior and environmental context. The system uses exterior camera data to identify relevant environmental features and feedback loops to refine gaze predictions, improving the reliability of attention assessment in dynamic driving scenarios.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4681955A1Driver monitoring system (DMS) for a vehicle
Publication Date: 2026.01.21 APTIV TECHNOLOGIES AG
  • EP4681955A1 patent drawingFigure 1A~1B
  • EP4681955A1 patent drawingFigure 2A~2B
  • EP4681955A1 patent drawingFigure 3

AI summary

The present disclosure relates to a Driver Monitoring System (DMS) for a vehicle, comprising at least one interior sensor configured to monitor an interior of the vehicle, at least one exterior sensor configured to monitor an environment of the vehicle, and a computing unit, wherein the computing unit is configured to retrieve, analyze and process sensor data from the interior sensor and the exterior sensor; identify a change of the perception in the sensor data from the interior sensor and/or the exterior sensor, wherein the change of the perception is within the field of view of an occupant of the vehicle; assign a virtual focus point to the identified change of the perception; and derive a prediction of the occupant's gaze vector based on the virtual focus point.