Driver State Detection Using DRTs in Monotonous Driving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for collecting data on human psychophysiological states during driving are prone to biases, making it difficult to derive reliable results for developing advanced driver assistance systems (ADAS), as they often introduce inconsistencies and limitations in data collection and analysis.
Innovation Solution
A method and system for collecting data involving a predetermined driving session with detection response tasks (DRTs) in a monotonous simulated environment, using physiological and behavioral data collection tools like EEG, ECG, and eye-tracking cameras, to assess driver drowsiness and cognitive load, with a structured protocol to minimize biases and ensure reliable data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is collected during real-life driving experiments to investigate human engagement to driving conditions, then the ecological validity of the data is improved, but biases are introduced that affect the reliability of the data
Solution Approach 1:
A simulator is introduced as an intermediary between the driver and the real driving environment. The simulator reproduces driving conditions in a controlled setting, allowing data collection without the uncontrolled variables and biases present in real-life experiments, thus maintaining data reliability while preserving ecological validity
Solution Approach 2:
Instead of collecting data in the original real-life driving environment, a copy or simulation of the driving environment is created. This copy replicates the essential characteristics of real driving conditions while eliminating the harmful biases, allowing reliable data collection on human engagement to driving conditions
2Quantity of substance
If the driving session duration is extended to collect sufficient psychophysiological data, then the quantity of data is improved, but driver fatigue and loss of data quality increase
Solution Approach 1:
The driving session is divided into periodic intervals with regular breaks. Data collection occurs in structured phases (driving segments followed by rest periods), allowing sufficient data to be gathered while preventing driver fatigue from compromising data quality. This periodic structure enables collection of multiple data points across different physiological states
Solution Approach 2:
The continuous driving session is segmented into discrete phases: driving segments for data collection and rest periods for driver recovery. This segmentation allows the total session to extend long enough to gather sufficient psychophysiological data while interspersing breaks that maintain data quality by preventing excessive fatigue
3Measurement precision
If multiple detection means are used to collect physiological data, then the measurement precision is improved, but the device complexity increases
Solution Approach 1:
Multiple detection means (ECG, EEG, eye-tracking, respiration sensors) are merged into an integrated data collection system. These separate measurement devices are combined and synchronized to collect physiological data simultaneously, improving measurement precision through multi-parameter monitoring while managing system complexity through unified data acquisition and processing architecture
Data Source
AI summary
It is herein disclosed a method of collecting data for determining a psychophysiological state of a human driver, comprising collecting data throughout a driving session, wherein the driving session comprises a predetermined time period of non-stop driving by a driver in a monotonous driving environment and a plurality of detection response tasks (DRTs) performed by the driver in the predetermined time period, wherein each DRT comprises the driver reacting to at least one visual stimulus; wherein said collecting data comprises: collecting, by a plurality of detection means, physiological data on the driver throughout the predetermined time period; wherein the physiological data collected comprises at least one of: heart rate data, electroencephalogram data, eye movement data, and respiration data; and collecting driver behaviour data, comprising recording steer wheel activity of the driver and recording a manner in which the driver reacts to the visual stimulus during the plurality of DRTs.