Driver Condition Estimating Device Using Head Movement Coherence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated driving systems lack effective early detection methods for driver abnormalities, such as illness or loss of ability to drive, which can lead to delayed emergency responses and increased safety risks.
Innovation Solution
A driver condition estimating device that measures head movement using a camera and calculates periodic feature amounts and coherence with lateral acceleration to detect signs of abnormality, employing nonlinear dimensionality reduction and machine learning for accurate classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If head movement measurement is used to detect driver abnormality, then detection capability is improved, but false detection occurs during cornering due to lateral acceleration
Solution Approach 1:
The patent introduces lateral acceleration as an intermediary variable to mediate between head movement and abnormality detection. By calculating coherence between head movement and lateral acceleration, the system distinguishes whether head movement is caused by vehicle cornering (high coherence) or driver abnormality (low coherence), thereby resolving false detections during cornering while maintaining detection accuracy.
2Measurement precision
If complex analysis methods are used to improve detection accuracy, then detection precision is improved, but computational complexity increases
Solution Approach 1:
The patent extracts only the essential features needed for detection: periodic feature amounts from head movement time series data and coherence values with lateral acceleration. By focusing on these specific extracted features rather than analyzing all raw data, the system achieves high detection precision while keeping computational complexity manageable through targeted feature extraction.
Data Source
AI summary
A driver condition estimating device includes circuitry configured to measure movement of the head of a driver output from a driver camera that photographs the driver and detect a sign of abnormality of the driver from the movement of the head. The circuitry determines existence of the sign of abnormality of the driver by calculating a periodic feature amount from time series data showing movement of the head of the driver, calculating coherence between the movement of the head of the driver and lateral acceleration acting on the head of the driver, calculating time series variation patterns from the obtained periodic feature amount and the obtained coherence, and comparing the obtained time series variation patterns with a predetermined threshold.