Risk Identification And Assisted Driving Strategies In Alpine Hypoxic Environments

Compared to the air oxygen environment in the plain,the hypoxic environment in the plateau area might has a negative impact on drivers’ physiological function and driving safety.It is of practical significance to identify the driving risks in the plateau area and propose a driving assistance strategy applicable to the plateau area to improve plateau highway road safety and to prevent traffic accidents.In this paper,by constructing a simulated hypoxic environment in the plain area,and conducting experimental study.The driving behavior,physiological signal data are collected,which provided data support for driver psychological load quantification and identify the driving risks.based on those models,proposing driving safety assistance measures and verifying the efficacy of the measures under the driving simulation experiment environment.First,a virtual simulation model was built based on UC-Win/Road and the driving simulation experiments was carried out.In order to study the differences between the plain and the plateau area,constructing a simulated hypoxic environment in the plain area by using the Everest Summit II altitude generator.two experiments are conducted in plain environment and simulated plateau hypoxic environment.Experimental data were acquired,including driving behavior and PPG.The subjective evaluation data of drivers were obtained through questionnaires.Secondly,using PPG to analyze the differences in drivers’ mental states between plains and plateau area,and build a model to quantify driver psychological load.Analysis the physiological characteristics and statistical differences of drivers between the two environments,and explore the causes and patterns of hypoxic environmental influences on drivers’ psychological states.The indicators of PPG were taken as the characterization parameters to construct the quantitative model of driver’s mental workload using exploratory factor analysis(EFA).Then the accuracy of the model was validated by the video recording of the experimental procedure,subjective evaluation data of drivers and frequency-domain analysis of the PPG indicators.Third,using driving behavior and PPG to analyze the differences in driving behavior characteristics between plains and plateau area,and build a model to identify the driving risks.Analysis the driving behavior characteristics and statistical differences of drivers between the two environments,and explore the causes and patterns of hypoxic environmental influences on driving behavior characteristics.The definition of driving risk types is proposed,and use the CNN-LSTM model to build a risk type identification model for driving based on driving behavior and PPG.Finally,the warning strategy were built from the perspective of driving risk state and driver’s psychological state,and experiments were conducted to verify the effectiveness of the real-time warning strategy.The CART was used to delineate the division interval between high and low risk of drivers’ mental state.A voice warning strategy was designed.To evaluate the effectiveness of the real-time warning strategy,the experiment was conducted again and subjective evaluations were obtained through questionnaires.