Driver Abnormality Detection via Dynamic Scene-Based Sensor Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driver abnormality detection systems face challenges in accurately determining driver abnormality in dynamic driving scenes, leading to noisy and impractical results, especially during cornering, where the accuracy of detecting autonomous periodicity of head rocking motion is low.
Innovation Solution
A driver abnormality determination system that includes circuitry to detect driving operations, head and eyeball behavior, and recognize driving scenes, using a sensor table to correlate detected values with driving scene data for precise abnormality determination, selecting body target parts with fewer noise elements per scene to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If driver abnormality is detected based on biological information from sensors mounted on the vehicle, then driver abnormality detection capability is provided, but determination accuracy becomes insufficient in actual driving scenes
Solution Approach 1:
The system dynamically selects and switches between different body target parts (head, eyeballs, steering wheel, seat) based on the current driving scene. This dynamic adaptation allows the system to maintain high detection accuracy across various driving conditions by using the most appropriate sensor data source for each specific scene, rather than relying on fixed sensor configurations.
Solution Approach 2:
The system changes the detection parameters by selecting different body target parts according to the driving scene. The circuitry determines which target part to monitor based on scene recognition, effectively changing the detection parameters to match the current driving conditions, thereby improving determination accuracy in actual driving scenes.
2Reliability
If autonomous periodicity of head rocking motion is detected during cornering, then driver abnormality detection is attempted, but noise increases and practicality decreases
Solution Approach 1:
The system dynamically adjusts the selected body target part based on the driving scene to avoid noisy detection conditions. During cornering scenes, the system switches from head motion detection to other target parts such as eyeball behavior or steering wheel operations, thereby eliminating the noise problem associated with head rocking motion during turns while maintaining abnormality detection capability.
Solution Approach 2:
The system uses driving scene recognition as an intermediary to determine the appropriate body target part for detection. This intermediary layer analyzes the current driving scene and selects the most suitable target part, acting as a mediator between the sensor data and the abnormality determination process to filter out noisy conditions.
3Ease of operation
If fixed body target parts are used for driver abnormality detection, then detection simplicity is maintained, but accuracy varies significantly across different driving scenes
Solution Approach 1:
The system achieves multi-functionality by using multiple body target parts for detection, where each target part serves as a universal detection source that can be selected based on the driving scene. This allows the system to maintain simplicity in operation while improving accuracy, as the circuitry automatically selects the appropriate target part without requiring complex manual configuration.
Solution Approach 2:
The system dynamically selects the body target part based on the current driving scene, transitioning from fixed to dynamic selection. This dynamic approach maintains operational simplicity through automated scene-based selection while significantly improving determination accuracy by adapting to the specific characteristics of each driving scene.
Data Source
AI summary
[Solution] A driver abnormality determination system includes circuitry configured to detect a driving operation of a driver, detect behavior of the driver's head and motion of eyeballs and/or movement of a sightline of the driver, recognize a driving scene in which the vehicle is driven; a memory that stores a sensor table of a relationship between the driving scene and detected values to be used to determine a driver abnormality; and determine the driver abnormality based on the sensor table, the driving scene, the driving operation and the behavior.