3D Eye Gaze Tracking for Driver Engagement Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing semi-autonomous vehicle systems fail to accurately determine driver engagement, leading to potential accidents due to inattentive driving, as they rely on user interactions and lack precise gaze detection.
Innovation Solution
A computing system uses three-dimensional eye gaze vectors to determine where a vehicle occupant is looking within the cabin, analyzing facial landmarks and applying machine learning to enhance precision, allowing for accurate identification of regions of interest and engagement levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional driver monitoring methods (steering wheel touch, eye open detection, blink speed) are used, then the system can detect basic driver states, but the measurement precision of gaze direction and engagement level is insufficient
Solution Approach 1:
The patent transitions from two-dimensional eye tracking to three-dimensional gaze vector detection by incorporating depth information from the distance sensor. This dimensional enhancement allows the system to calculate precise gaze directions in 3D space, resolving the measurement precision issue while maintaining manageable system complexity through modular sensor integration.
Solution Approach 2:
The patent introduces a distance sensor as an intermediary component that provides depth information to the computing system. This intermediary enables the calculation of three-dimensional gaze vectors by combining distance data with eye position data, achieving high measurement precision without requiring direct complex eye-tracking hardware.
2Measurement precision
If the system monitors driver gaze continuously with high precision, then driver engagement can be accurately assessed, but the use of energy increases due to multiple sensors and processing requirements
Solution Approach 1:
The patent implements selective monitoring by determining whether the vehicle is in autonomous or manual mode and only performing high-precision gaze analysis when necessary (e.g., during autonomous operation or when distraction is suspected). This partial action approach maintains safety requirements while reducing overall energy consumption compared to continuous high-precision monitoring.
Solution Approach 2:
The computing system performs multiple functions using the same sensor data: it monitors driver engagement, detects distractions, determines gaze direction, and assesses safety conditions. This multi-functionality allows the system to achieve high measurement precision for driver engagement while consolidating processing tasks to reduce energy consumption.
3Measurement precision
If the system uses multiple sensors and machine learning algorithms to determine gaze vectors, then the measurement precision improves, but the device complexity increases
Solution Approach 1:
The patent divides the monitoring system into separate functional modules: eye position detection, distance measurement, gaze vector calculation, and engagement assessment. This segmentation allows each component to be optimized independently while maintaining overall measurement precision, reducing the perceived complexity through modular architecture.
Solution Approach 2:
The computing system automatically processes the sensor data using machine learning algorithms to calculate gaze vectors and assess driver engagement without requiring manual calibration or intervention. This self-service capability maintains high measurement precision while reducing the operational complexity for users.
4Reliability
If the system restricts driver interaction with the head unit based on gaze detection, then safety is improved, but the ease of operation decreases due to limited access
Solution Approach 1:
The patent implements dynamic access control where the restriction level adjusts based on real-time gaze detection. When the driver's gaze is detected on the road, full access to the head unit is permitted. When gaze shifts to side windows or other distractions are detected, access is restricted. This dynamic approach maintains safety while preserving ease of operation when appropriate.
Solution Approach 2:
The system provides feedback to the driver about their engagement level and restricts head unit access based on detected distraction behaviors. This feedback mechanism encourages safe driving habits while allowing easy access to controls when the driver is properly engaged, balancing safety and operability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
According to the techniques of this disclosure, a method includes capturing, using a camera system of a vehicle, at least one image of an occupant of the vehicle, determining, based on the at least one image of the occupant, a location of one or more eyes of the occupant within the vehicle, and determining, based on the at least one image of the occupant, an eye gaze vector. The method may also include determining, based on the eye gaze vector, the location of the one or more eyes of the occupant, and a vehicle data file of the vehicle, a region of interest from a plurality of regions of interests of the vehicle at which the occupant is looking, wherein the vehicle data file specifies respective locations of each of the plurality of regions of interest, and selectively performing, based on the region of interest, an action.