Driver’s Mental Workload Evaluation And Application Based On Heart Rate Variability

Human error is a main factor of traffic accidents. Whereas, unsuitable mental load level of driver is an important factor to impel people make error performing driving duties in the complex transportation system consist of driver, vehicle, and transportation environment. Increased visual information of driver would cause the increase of mental workload. Driver prone to entering a state of overloading or ultra-low mental workload affected by the information quantity, the complexity of information, and the permissible time for operating the driving-related tasks and non-driving-related tasks, which will result the slowly or error decisions, causing the unsafe driving behavior such as driving errors. Therefore, the driver’s mental load become important research problem in the field of transport ergonomics. This paper mainly studied the driver with the mental load evaluation menthod based on heart rate variability, through multivariate statistical process control method to judge the driver’s mental workload level. Then study the accuracy of the pulse signal having a sensor for detecting heart rate variability analysis, validate model can be used to evaluate the usefulness of the wearable device theoretically.Firstly, this paper researched the current situation of Mental Workload on efficiency scholars conducted widely, evaluation methods in previous studies on the basis of mental workload for the research was designed for different levels of mental workload experimental test program. Experiment platform selection ErgoLAB human-computer synchronization experimental platform in the city to carry out the actual vehicle road test, designed and completed 10 experiments with the same complexity of experimental route different road sections, the recording of the 10 drivers with the corresponding experimental ECG scene, according to experiment number experiment data storage.Thirdly, in this paper, the experimental data were ECG signal preprocessing, R wave peak identification, extraction RR interval, RR interval spline interpolation, etc. obtained RR interval time series interval through multiple sets of experimental data testing, made a very good RR interval extraction, to ensure the accuracy of the analysis of HRV data source. Using multivariate statistical process control variables, heart rate variability in the frequency domain sample index variable, build pilot mental workload evaluation model. Multiple sets of experimental data analysis found that the model for the next complicated road driver mental workload level monitoring with a strong sensitivity to changes in multiple sets of data validation Q value can be used as mental workload evaluation of reliability indices.Finally, after verifying the driver’s heart rate variability for having mental load evaluating reliability based on paper studies the photoelectric sensor having a pulse detection signal for heart rate variability analysis accuracy, and enforceable by multiple sets of data, the results show that the pulse signal detection photoelectric sensor for analysis of heart rate variability compared with ECG accuracy rate of more than 95%. The accuracy of the analysis show that the proposed load evaluation model applications in wearable auxiliary driving device is large, verify the practicality of this approach.