Driver State Estimation Using Adaptive Sensor Selection in Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle systems face challenges in distinguishing between a sleeping driver and an abnormal driver state, particularly in automated driving modes where the driver is permitted or not permitted to sleep, leading to inefficiencies and increased processing requirements.
Innovation Solution
A vehicle device and method that utilize a combination of sensors to estimate driver states, adjusting the types of sensors used based on the automation level, reducing unnecessary processes by using fewer sensors during sleep-permitted automated driving while ensuring accurate differentiation between sleeping and abnormal states during sleep-unpermitted driving.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple types of sensors are used to estimate driver state, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system dynamically adjusts the sensor configuration based on the determined driver state. When the driver is determined to be in a sleeping state during sleep-permitted automated driving, the system reduces the number of sensors activated. When the driver is determined to be in an abnormal state or during sleep-unpermitted driving, the system activates all necessary sensors. This dynamic adaptation resolves the contradiction by making the device complexity variable rather than fixed.
Solution Approach 2:
Different sensor types are selectively applied based on the specific driving context and driver state. The system uses a plurality of sensor types (camera, microphone, seat sensor, steering wheel sensor, biological sensor) but only activates the necessary subset for each situation. This local quality approach ensures high measurement precision when needed while reducing overall device complexity by not continuously using all sensors.
2Reliability
If multiple types of sensors are continuously used, then reliability is improved, but loss of energy increases
Solution Approach 1:
The system performs periodic determination of driver state and automatically adjusts sensor activation accordingly. During sleep-permitted automated driving, when the driver is determined to be sleeping, the system periodically monitors using reduced sensors. During sleep-unpermitted driving or when abnormal states are detected, the system switches to continuous monitoring with all sensors. This periodic reassessment and adaptive activation maintains reliability while significantly reducing energy loss during periods when full monitoring is not required.
3Productivity
If sensor usage is reduced during sleep-permitted automated driving, then productivity is improved, but measurement precision may deteriorate
Solution Approach 1:
The system performs preliminary determination of the driver state using available sensors before reducing sensor usage. When the driver is determined to be in a sleeping state during sleep-permitted automated driving, the system confidently reduces sensors because the sleeping state has already been established. This preliminary action ensures that measurement precision is maintained during the initial detection phase, allowing productivity to improve during the subsequent reduced-monitoring phase without sacrificing overall detection accuracy.
Data Source
AI summary
A vehicle device or a vehicle estimation method estimates whether the driver is in an abnormal state different from a sleeping state by using a plurality of types of sensors, determines whether the vehicle is in a sleep-permitted automated driving or a sleep-unpermitted driving. When determining that the vehicle is in the sleep-unpermitted driving, the device or the method estimates whether the driver is in the abnormal state by using the plurality of types of the sensors. When determining that the vehicle is in the sleep-permitted automated driving, the device or the method reduces the plurality of types for estimation.