Vehicle Occupant Gaze Detection for Autonomous Mode Transition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Determining the direction an occupant is looking during autonomous vehicle operation is challenging, as they may not maintain focus on the roadway, posing a problem for the vehicle's computer in transitioning between manual and autonomous control modes effectively.
Innovation Solution
A system using a computer with a processor and memory that determines the occupant's gaze direction by analyzing image data, employing a machine learning program to identify landmarks and calculate probabilities of gaze points, and suppressing manual control of vehicle components when the gaze distance exceeds a threshold, allowing transition to autonomous mode.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the occupant looks away from the roadway during autonomous operation, then the occupant can relax attention, but the vehicle computer cannot reliably determine when to transition between manual and autonomous control modes
Solution Approach 1:
The patent replaces manual monitoring of occupant attention with an automated computer vision system using cameras and machine learning algorithms to detect gaze direction. The system captures images of the occupant's face, identifies eye and pupil positions, and calculates gaze direction automatically, eliminating the need for mechanical or manual attention monitoring while ensuring reliable control mode transitions.
2Measurement precision
If the system uses image analysis and machine learning to determine gaze direction, then gaze detection accuracy is improved, but computational complexity and processing requirements increase
Solution Approach 1:
The patent segments the gaze detection task into distinct processing stages: face detection, eye localization, pupil position identification, and gaze direction calculation. Each stage processes specific features independently, reducing computational complexity at each step while maintaining overall detection accuracy. The machine learning model is trained to recognize specific facial landmarks and eye characteristics in sequence.
Solution Approach 2:
The system performs preliminary actions by pre-training the machine learning model with large datasets of facial images and gaze patterns before deployment. The model learns to identify facial landmarks, eye positions, and pupil orientations in advance, enabling rapid real-time gaze detection without requiring complex computations during actual vehicle operation. Pre-computed facial feature databases are used to accelerate processing.
Data Source
AI summary
A computer includes a processor and a memory, the memory storing instructions executable by the processor to determine respective probabilities of a direction of a gaze of a vehicle occupant toward each of a plurality of points in an image, determine a gaze distance from a center of the image based on the probabilities, and, upon determining that the gaze distance exceeds a threshold, suppress manual control of at least one vehicle component.