Опубликована: Янв. 1, 2024
Язык: Английский
Опубликована: Янв. 1, 2024
Язык: Английский
Future Transportation, Год журнала: 2024, Номер 4(1), С. 283 - 298
Опубликована: Март 13, 2024
This paper summarizes the research on countermeasures against driver fatigue based a comprehensive systematic literature review. Driver fatigue, induced by task monotony during conditional automated driving (CAD, SAE Level 3), can increase risk of road accidents. There are several measures that counteract and aim to reduce caused fatigued in context CAD. Twelve selected articles focusing CAD were analyzed. The findings conclusions presented, themselves their implementation. critically discussed, especially regarding effectiveness applicability. They seem be effective counteracting fatigue. However, not easily compared because they studied various experimental settings measurements used. Different have proven reducing For this reason, further investigation is needed gain insights into applications, advantages, disadvantages. Further studies will conducted verify best solution
Язык: Английский
Процитировано
1Traffic Injury Prevention, Год журнала: 2024, Номер unknown, С. 1 - 8
Опубликована: Июнь 11, 2024
Objective Vehicle automation technologies have the potential to address mobility needs of older adults. However, age-related cognitive declines may pose new challenges for drivers when they are required take back or "takeover" control their automated vehicle. This study aims explore impact age on takeover performance under partially driving conditions and interaction effect between voluntary non-driving-related tasks (NDRTs) performance.
Язык: Английский
Процитировано
1Journal of Safety Research, Год журнала: 2024, Номер 91, С. 314 - 325
Опубликована: Окт. 4, 2024
Язык: Английский
Процитировано
1Transportation Research Part F Traffic Psychology and Behaviour, Год журнала: 2024, Номер 103, С. 18 - 34
Опубликована: Апрель 4, 2024
Язык: Английский
Процитировано
0Опубликована: Янв. 1, 2024
Driver fatigue continues to be a major contributor road traffic accidents, significantly compromising driver's ability safely operate vehicle. Existing driving detection methods rely on eyelid closure, vehicle information, or physiological parameter detection. However, each individual feature method has limitations, which in turn affects the accuracy of and efficiency prediction. This paper proposes novel approach. It establishes model using facial features, head posture, PPG signals. We conducted real-road experiments gathered these sources characteristic signal data from sample 30 drivers. Using locating 68 key points, we extracted features parameters, while heart rate variability parameters were derived signal. The 5-dimensional dataset was established by fusing above 10-dimensional parameters. Construct long short-term memory network (LSTM) model, use four algorithms, Momentum, RMSProp, Adam, SGD, optimize model. 2880 are used as training set, 720 test results optimal effect Adam-optimized LSTM with 97.36% accuracy, 97.4% precision, recall, 0.97 F1 . shows that is able provide timely accurate prediction warning for drivers who already fatigued.
Язык: Английский
Процитировано
0Опубликована: Янв. 1, 2024
Язык: Английский
Процитировано
0