Driver Gaze-Aware Object Detection for AEB False Trigger Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current driver monitoring systems face limitations in accurately determining driver awareness, leading to false triggers in assistance systems like Automatic Emergency Brake (AEB), as they rely solely on driver input and do not account for the driver's field of view and gaze direction.
Innovation Solution
A computer-implemented method that uses an object tracking system and a driver monitoring camera to determine the driver's field of view and gaze direction, allowing for the detection of objects within or outside the driver's view, and adjusts detection values based on object position and movement, thereby providing accurate alerts and compensation measures to prevent false positives and improve driver assistance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If driver awareness is determined solely based on driver input signals, then the system is simple to implement, but false triggers occur in assistance systems like AEB
Solution Approach 1:
The patent introduces an intermediary verification mechanism by comparing object detection data with driver field of view information. This intermediary check determines whether the driver is actually aware of detected objects, serving as a mediator between object detection and assistance system triggering, thereby reducing false positives without requiring complete system redesign
Solution Approach 2:
The system implements feedback by continuously monitoring driver gaze direction and field of view, then using this information to adjust the triggering conditions of assistance systems. The feedback loop compares driver awareness state with detected object positions, dynamically adjusting system behavior to avoid false triggers while maintaining safety
2Reliability
If the system triggers warnings or emergency brake activation earlier to avoid collisions, then collision avoidance improves, but false triggers increase
Solution Approach 1:
The patent applies local quality by differentiating the triggering conditions based on the specific spatial relationship between detected objects and the driver's field of view. Objects within the driver's field of view receive different treatment (later triggering assuming driver awareness) compared to objects outside the field of view (earlier triggering), creating localized quality in the safety response
3Measurement precision
If the system monitors driver gaze direction and field of view to improve accuracy, then false positives are reduced, but system complexity increases
Solution Approach 1:
The patent achieves universality by using the driver monitoring camera for multiple functions: it tracks driver position, determines gaze direction, calculates field of view boundaries, and provides this information to the assistance system. This multi-functional approach reduces the need for separate dedicated components for each measurement, thereby limiting the increase in system complexity
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
The present invention concerns a computer-implemented method for assisting the driver of a vehicle. This is achieved by determining, based on a gaze direction of the driver, a field of view of the driver; detecting, using an object tracking system, that one or more objects are present around the vehicle; determining that one or more of the detected objects are outside of the field of view of the driver; and performing an action if it is determined that one or more objects are outside of the field of view of the driver. The invention also describes a driver assistance system, a program for performing the method for assisting the driver, and a vehicle having the driver assistance system.