Driver State Estimation Using Saccades and Attention Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional driver state estimation techniques inaccurately classify inattentive states due to factors like disease or aging, failing to distinguish them from other abnormal states.
Innovation Solution
A driver state estimation apparatus that utilizes travel environment information and line-of-sight detection to calculate an inattentive probability using feature values like saccade frequency and amplitude, top-down attention score, and corrects these values based on road gradient, curvature, illuminance, and vehicle speed to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional line-of-sight detection technique is used to estimate driver state, then the system can detect saccade movement, but it cannot distinguish inattentive states from abnormal states caused by disease or aging
Solution Approach 1:
The patent segments the driver state estimation into multiple independent indicators: saccade frequency, saccade amplitude, and top-down attention score. Each indicator is evaluated separately and then integrated, allowing the system to distinguish between temporary inattentive states and persistent abnormal states caused by disease or aging through pattern recognition across multiple dimensions.
Solution Approach 2:
The patent introduces dynamic parameter changes by incorporating travel environment information (road gradient, curvature, illuminance, vehicle speed) to adjust the evaluation criteria. This allows the system to adapt to varying driving conditions and distinguish genuine inattentive states from environmental influences, improving both measurement precision and reliability.
2Device complexity
If the system focuses on single indicator like saccade frequency, then the detection is simple, but the estimation accuracy decreases when driver's line of sight movement changes due to disease or aging
Solution Approach 1:
The patent merges multiple indicators (saccade frequency, saccade amplitude, top-down attention score) into a comprehensive evaluation system. By combining these indicators with travel environment information, the system achieves high measurement precision while maintaining reasonable complexity through integrated processing.
Solution Approach 2:
The patent creates a multi-functional evaluation system that can assess both normal inattentive states and abnormal states caused by disease or aging. The system universally applies to various driving conditions by incorporating environmental factors, making it adaptable to different driver states without requiring separate detection methods.
3Device complexity
If the system does not consider travel environment information, then the processing is simpler, but the estimation becomes inaccurate due to environmental factors affecting line-of-sight behavior
Solution Approach 1:
The patent applies preliminary action by pre-acquiring and storing travel environment information (road gradient, curvature, illuminance, vehicle speed) before driver state estimation. This pre-prepared environmental data is then integrated with line-of-sight detection results, improving estimation precision without significantly increasing processing complexity during actual driver monitoring.
Data Source
AI summary
A driver state estimation apparatus includes a controller configured to estimate whether the driver is in a first state based on travel environment information and the driver's line of sight. Based on the travel environment information and the driver's line of sight. The controller acquires feature values xi including a frequency and an amplitude of a saccade of the driver and a top-down attention score indicating a degree of deviation from appropriate line-of-sight distribution to an attention object around the vehicle, uses the acquired feature values xi and a preset weight coefficient ai for each of the feature values xi, to calculate a first probability p representing a probability that the driver is in the first state. The controller estimates that the driver is in a first state when the calculated first probability p is equal to or higher than a predetermined value continues for a predetermined time or longer.