Vehicle Map Updating Using Driver Attention Regions

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing high-definition (HD) maps lack integration with driver monitoring system (DMS) data, particularly gaze and head pose estimation, which are crucial for enhancing situational awareness and safety in autonomous driving and driver assistance systems, especially in complex scenarios.

Innovation Solution

A method to integrate DMS data, including gaze and head pose estimation, with HD maps to identify and mark 'attention regions' or 'anomaly events' on the map, using sensors to detect driver behavior and environmental factors, and update the map with these regions to enhance safety and navigation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If DMS data integration is implemented to enhance map accuracy and safety, then map utility and safety features are improved, but device complexity and data processing requirements increase

Engineering Contradiction:
Improvemap safety featuresVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent combines DMS data (gaze estimation, head pose, driver state) with existing HD map data and external sensor data into a unified map representation. This merging allows the system to leverage existing infrastructure while adding driver-centric safety information, resolving the contradiction by integrating multiple data sources into a single enhanced map structure that improves reliability without requiring entirely separate systems

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The enhanced map serves multiple functions: traditional navigation, safety hazard identification, driver behavior analysis, and route optimization. By making the map multi-functional, the system justifies the increased complexity through diverse utility, allowing a single system to address multiple safety and navigation requirements simultaneously

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

2Loss of information

If gaze and head pose estimation data are integrated into HD maps, then situational awareness is enhanced, but data processing complexity and computational requirements increase

Engineering Contradiction:
Improvesituational awarenessVSAvoiddata processing complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The patent extracts specific relevant features from DMS data (gaze direction, head pose angles, attention regions) and separates them from the raw sensor streams. By extracting only the essential driver state information needed for map enhancement, the system reduces processing complexity while maintaining situational awareness benefits, avoiding the need to process entire DMS data sets

Inventive Principle:
Principle #2Taking out (Extraction)

3Measurement precision

If real-time DMS sensor data is used to update maps, then map accuracy is improved, but data acquisition and processing time increase

Engineering Contradiction:
Improvemap accuracyVSAvoiddata processing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent implements partial updates to the map by only modifying regions where driver attention data indicates hazards or points of interest, rather than processing and updating the entire map continuously. This selective updating approach maintains high map accuracy in critical areas while significantly reducing overall data processing time and computational burden

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP4671685A1A method of updating a map in a vehicle
Publication Date: 2025.12.31 APTIV TECHNOLOGIES AG
  • EP4671685A1 patent drawingFigure 1
  • EP4671685A1 patent drawingFigure 2
  • EP4671685A1 patent drawingFigure 3

AI summary

A method of using sensor inputs from a driver monitoring system (DMS) to update information on a map (M), comprises the steps of (i) acquiring sensor data from sensors in a vehicle or a fleet (F), at least including an internal sensor (10, 11, 12, 13) configured for monitoring a driver characteristic; and (ii) identifying an extraordinary event (e.g. sudden braking, rapid head movement or increased stress level) from the data along with a location (e.g. by GPS) of where the event occurred. A global map (M) can then be updated at a server (50) to show the location of the extraordinary event as an attention region; i.e. an area where heightened awareness may be needed by a driver or autonomous vehicle. In a particular form, the attention region is determined/dimensioned according to a gaze pattern of a driver when the extraordinary event took place.