Gaze Target Detector Using Dynamic Threshold Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing gaze target detection systems for vehicles, such as those disclosed in JP-A Nos. 2008-018853 and 2013-255168, are inadequate in detecting occupant gaze in high-speed environments and fail to accurately specify target objects when the vehicle is turning and the target is moving, as they do not effectively utilize vehicle traveling data like speed and curvature.
Innovation Solution
A gaze target detector system that includes a line-of-sight detector, relative speed data acquiring unit, relative position data acquiring unit, curvature calculator, and threshold adjuster, which calculates a variable threshold based on relative speed, position, and curvature to determine the gaze target, allowing for accurate detection of the occupant's gaze in various traveling environments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a fixed threshold is used to determine gaze target, then the detection method is simple, but the detection accuracy deteriorates in high-speed and turning conditions
Solution Approach 1:
The patent applies the dynamics principle by making the threshold variable instead of fixed. The threshold is dynamically adjusted based on vehicle traveling state (speed and curvature) to adapt to different driving conditions. This resolves the contradiction by allowing the system to maintain simple operation while achieving high detection accuracy through context-aware threshold adaptation.
Solution Approach 2:
The patent changes the parameter of the threshold from a constant value to a variable that depends on vehicle speed and curvature. By modifying this key parameter based on traveling conditions, the system achieves accurate gaze target detection in diverse scenarios without significantly increasing overall system complexity.
2Device complexity
If vehicle traveling data is not utilized, then the detection system is simple, but the ability to accurately specify target objects in turning conditions deteriorates
Solution Approach 1:
The patent makes the detection system multi-functional by having it process both line-of-sight data and vehicle traveling data (speed and curvature). This universal approach allows the same system to accurately specify target objects whether the vehicle is stationary, moving straight, or turning, thereby improving reliability without requiring separate specialized systems.
Solution Approach 2:
The patent implements feedback by using vehicle traveling state information to adjust the gaze target detection threshold. The system continuously monitors vehicle speed and curvature, and uses this feedback to adaptively modify detection parameters, ensuring reliable target specification under varying driving conditions.
3Ease of operation
If a fixed threshold is used, then the system operation is simple, but the adaptability to different traveling environments deteriorates
Solution Approach 1:
The patent applies dynamics by enabling the threshold to automatically adapt to different traveling environments based on vehicle speed and curvature data. This dynamic adjustment mechanism maintains ease of operation while significantly improving adaptability across diverse driving conditions such as high-speed travel, low-speed maneuvering, and turning.
Solution Approach 2:
The patent changes the threshold parameter based on detected traveling conditions, allowing the system to adapt to different environments without requiring manual reconfiguration. This parameter adaptation maintains operational simplicity while achieving environment-specific optimization.
Data Source
AI summary
A gaze target detector includes a line-of-sight detector, a relative speed data acquiring unit, a relative position data acquiring unit, a curvature calculator, a threshold adjuster, and a gaze determination unit. The line-of-sight detector detects the line of sight of an occupant in a vehicle. The relative speed data acquiring unit acquires a relative speed between the vehicle and a gaze target at which the occupant is gazing. The relative position data acquiring unit acquires a relative position between the vehicle and the gaze target. The curvature calculator calculates the curvature of a traveling track of the vehicle. The threshold adjuster adjusts a threshold based on at least one of the relative speed, the relative position, or the curvature. The threshold is used to determine the gaze target. The gaze determination unit determines the gaze target as a gaze event based on the threshold adjusted.