Driver Attention Research Articles

Looking out or Looking Away?—Exploring the Impact of Driving With a Passenger on Young Drivers’ Eye Glance Behavior

Objective To explore how passenger presence and the degree of association between young driver and passenger influences young drivers’ eye glance behavior when they are subjected to distraction. Background Young drivers (18–20 years old) are at an elevated crash risk when subjected to distraction. They are likely to be distracted even further when they drive with passengers. However, the eye glance behavior of these drivers when driving with passengers has not been explored. Method Eye glance data of 34 young drivers between the ages of 18 and 20 years were collected. Participants drove with and without a passenger while subjected to three distracting tasks (visual-manual, cognitive, or visual-cognitive) and driving scenarios that required driver attention. Results Visual-cognitive as well as visual-manual states of distraction result in higher mean and standard deviation of glance duration, along with higher number of glances away from road. Passenger presence is found to negatively influence young drivers’ eye glance behavior. The degree of association between the young driver and the passenger may help reduce the deviation of eye glances towards the task-related objects. Conclusion In addition to distraction, passengers have a negative influence on the eye glance behavior of young drivers. However, a high degree of association between driver and passenger may mitigate the negative impact of distraction on the eye glance behavior of young drivers. Application (non-theoretical works) This research may aid in the design of interventions that improve young drivers’ eye glance behavior when they drive with their peers.

Driver's Visual Attention Characteristics and Their Emotional Influencing Mechanism under Different Cognitive Tasks.

The visual attention system is the gateway to the human information processing system, and emotion is an important part of the human perceptual system. In this paper, the driver’s visual attention characteristics and the influences of typical driving emotions on those were explored through analyzing driver’s fixation time and identification accuracy to different visual cognitive tasks during driving. The results showed that: the increasing complexity of the cognitive object led to the improvement of visual identification speed. The memory and recall process increased drivers’ fixation time to cognitive objects, and the recall accuracy decreased with the increase in time interval. The increase in the number of cognitive objects resulted in the driver improving the visual identification speed for the cognitive object at the end of the sequence consciously. The results also showed that: the visual cognitive efficiency was improved in the emotional states of anger and contempt, and was decreased in the emotional states of surprise, fear, anxiety, helplessness and pleasure, and the emotional state of relief had no significant effect on the visual cognitive efficiency. The findings reveal the driver’s visual information processing mechanism to a certain extent, which are of great significance to understand the inner micro-psychology of driver’s cognition.

Analysis of Driver Gaze and Attention to Traffic Signs

A driver’s actions and intent can be factors in enabling advance driver assistance systems (ADASs) to assist drivers and avoid accidents. A driver’s gaze can provide insight into the driver’s intent or awareness of situations. Knowing that a driver gazed at a traffic sign or missed a traffic could provide indications of whether the driver is alert to impending changes in the driving environment, such as curves and stop signs. For ADASs to determine the importance of a driver seeing or missing a sign, it is important to understand the driving environment and situation. A first step is to understand what signs drivers do see or miss while driving. This contribution presents the results of analyzing driving sequences to assess traffic signs that drivers may or may not have gazed upon. The results suggest that drivers may miss 20% of traffic signs though the percentage varies depending on the type of sign. The analysis uses image sequences of the driving environment and gazes data captured during driving. The methods used in our analysis included determining whether a driver’s gaze has fallen on the image of a traffic sign or not and subsequently determining signs missed during driving. The methods presented can be useful in other scenarios involving the analysis of driver gaze and have implications for the design of future ADASs and for understanding of driver gaze and awareness.

ДОСЛІДЖЕННЯ СВІТЛОПОВЕРТАЛЬНИХ ТА КОЛОРИМЕТРИЧНИХ ХАРАКТЕРИСТИК ДОРОЖНІХ ЗНАКІВ В ЗАЛЕЖНОСТІ ВІД ТЕХНОЛОГІЇ ЇХ ВИГОТОВЛЕННЯ

Introduction. Road signs form the framework of the driver's decision-making system in the process of vehicle driving, at the junction, separation and intersection of traffic and pedestrian flows. The most important in road signs are their luminance factor and retro reflectivity at nighttime; signs must have clear feedback. The more fully and clearly driver is informed about the conditions and needed modes of traffic,the more accurate and error-free are his actions. Problem statement. According to the analysis of accident rate statistics, most traffic accidents occur due to the inability of the driver to evaluate adequately the situation in conditions of insufficient visibility of road markings, signs and unsatisfactory lighting. According to researches, it has been determined, that improving the visibility of markings and signs can reduce traffic accidents by 10-30%. According to statistics, about 27% of traffic accidents occur due to driver inattention or loss of concentration. In Ukraine, such studies are almost non-existent, but according to traffic police data, in traffic accidents in the cities 50% of drivers consider also the cause of the collision tiredness or loss of concentration. An urgent task to eliminate the factors mentioned above is the use of high-quality informative road signs that will draw the driver's attention to the elements of the road situation (the presence of pedestrian crossings, possible complications of traffic conditions, trajectory of own vehicle move, etc.). For mounting quality signs, first of all, it is need to follow all the rules of technology of their production. Purpose. To study the retroreflective and colorimetric characteristics of road signs depending on the technology of their manufacture. Мaterials and methods. Laboratory studies for determination the influence of the technology of signs manufacturing on the specific luminous intensity factor, chromaticity coordinates, luminance factor. Results. The results of laboratory tests of road signs samples manufactured by different technologies for determination of specific luminous intensity factor, chromaticity coordinates, luminance factor are presented. Conclusions. Based on the results of studies, it can be concluded that the manufacturers of road signs need to work very carefully on the selection of inks for digital printing in order to achieve an image with high color endurance. Manufacturers are advised to have the needed devices to control the color at the stage of manufacture. Also, it can be noted that the specific luminous intensity factor of the road sign image of non-white characters made by digital printing is less important than the signs made by other technologies. Manufacturers need to draw attention to this and work over the possibility of achieving higher values of retro reflectivity. Keywords: road sign, lumina

Smart Diagnosis: Deep Learning Boosted Driver Inattention Detection and Abnormal Driving Prediction

Inattentive driving is one of the high-risk factors that causes a large number of traffic accidents every year. In this article, we aim to detect driver inattention leveraging on large-scale vehicle trajectory data while at the same time explore how do these inattentive events affect driver behaviors and what following reactions they may cause, especially, for commercial vehicles. Specifically, the proposed system targets four most commonly occurring critical inattentive events, including smoking, phone call, turning back, and yawning. By applying a deep convolutional neural network (CNN) (Inception v3) with two data augmentation routines—Mixup and synthetic minority oversampling technique (Smote), we are able to balance the training data distribution and improve the generalization of the classification model. Then, based on the output derived from the inattention detection combining with point of interest (POI) and climate data, a long short-term memory (LSTM)-based model is deployed to predict driver upcoming abnormal operations on road (due to inattention) which may result in potential dangerous driving conditions, such as sudden acceleration/deceleration, aggressive left/right lane change, etc. To evaluate our proposed system, we collect more than 120000 real-world driving traces from over 200 drivers. The experimental results show that our model achieves a weight accuracy (WA) of 92.27% for inattentive driving detection and a WA of 91.67% for abnormal driving prediction, demonstrating its great potential of shaping good driving habits and promoting road safety.

Inclusive ethical AI in human–computer interaction in autonomous vehicles

Artificial intelligence (AI) used in autonomous vehicles (AV) to check for driver alertness is a critical piece of technology that makes the decision to hand over control to the human if there is a disengagement of the autonomous capability. It is important that this AI be inclusive without bias, because treating drivers differently will have an impact on the safety of humans not only in the vehicle but also on the road. This paper evaluates the AI that powers driver attention systems in the car to check if the AI treats all humans inclusively the same way beyond their ethnicity, gender and age, and whether it follows AI ethics principles of trust, transparency and fairness. Driver attention is built using two different AI models. One uses camera data to recognise humans and the other evaluates whether the human is alert. We found that both these AI models are biased and not inclusive of all people in all situations. We also found that there are unethical practices in how humans are tracked to check for alertness by using infrared sensors that track their retina movements without any concept of consent or privacy of people being tracked in the vehicles. The paper builds upon prior research on face detection outside the car and research that shows that car cognition AI does not recognise all humans on the road equally. We present research results about how the car is biased against some humans in its face identification and how the assertion of alertness of humans to hand over control during an emergency is fundamentally flawed in its definition of alertness. We recommend mitigation techniques and call for further research to build upon our work to make the AV inclusive with bias mitigation of bias in all forms of AI in AVs.

Pedestrian fatalities in darkness: What do we know, and what can be done?

An alarming, consistent increase in U.S. pedestrian fatalities since 2009 culminated in a 28-year high of 6,283 pedestrians killed in 2018. Yet these numbers obscure a second alarming trend: 75% of pedestrian fatalities occur in darkness, and nearly 90% of the increase in pedestrian fatalities from 2009 to 2018 occurred in darkness.This paper examines data on pedestrian fatalities at the national level and pedestrian fatalities and serious injuries in California from 2012 to 2017 to better understand correlates of severe pedestrian injuries in darkness. Binary and multinomial logit models reveal that variables related to roadway design and operations (e.g., speed limits, number of lanes, roadway type, and presence of traffic control) – but not speeding – are significantly associated with the likelihood of a pedestrian fatality or serious injury occurring in darkness as compared to daylight. Critically, these factors – which were consistent for fatalities regardless of lighting presence and roadway type, with few exceptions – are all worse in darkness because they are negatively affected by a lack of visibility. Alcohol usage by drivers or pedestrians and sociodemographic characteristics were also positively associated with severe injuries in darkness.Our findings urge an explicit consideration of pedestrian safety in darkness in all future design and retrofit decisions, and particularly on higher-speed, multi-lane roadways. Immediate solutions include roadway designs and policies that slow drivers, particularly at night, and that increase illumination and driver attention, such as through additional roadway lighting, high-visibility countermeasures and protected crossings, and adaptive lighting and detection technology for vehicles.

Analysis of Driver Behavior and Intervehicular Collision: A Data-Based Traffic Modeling and Simulation Approach

The emergence of intelligent connected vehicles (ICVs) is expected to contribute to resolving traffic congestion and safety problems; however, it is inevitable that ICV safety issues in mixed traffic (involving ICVs and human driven vehicles) will be a critical challenge. The numerical simulation of scenarios involving a mix of different driving profiles is expected to be an important safety assessment tool in the process of testing and validating ICVs, especially regarding extreme scenarios, including car collisions, which are rarely captured in real-world datasets. In this study, we propose a novel approach for car collision generation in numerical simulations based on the assumption that car collision occurrences are mostly associated with certain specific driver profiles. Using a dataset provided by the Next Generation Simulation (NGSIM) project, NGSIM 101 dataset, we identify three different driver profiles: aggressive, inattentive, and normal drivers. We then replicate car collision occurrences by varying the percentages of these three driver profiles in the simulated environment, allowing us to establish a relationship between driver profiles and car collision occurrences. We also investigate the severity of car collisions and classify them with respect to the driver profiles of the cars involved in the collisions. Our approach of replicating car collision occurrences in numerical simulations will facilitate the testing and validation of ICVs in the future, especially regarding the testing of ICV functionalities in dealing with traffic accidents.

Head pose estimation: An extensive survey on recent techniques and applications

Biometric based systems are involved in many areas, from surveillance to user authentication, from autonomous systems to human-robot interactions. Head pose estimation (HPE) is the task to support biometric systems in which any of the biometric traits of the head is involved, as face, ear or iris. This particular biometric branch finds its application in driver attention detection, surveillance for recognition, face frontalization, best frame selection and so on. The goal of HPE is to determine the head pose orientation (yaw, pitch, roll). The implemented methods use different techniques depending on the kind of input. In this survey we present an overview of involved datasets, recent techniques and applications. We evaluate and compare the different approaches with respect to their advantages and practical usage. In addition, we propose a technical comparison between training and training-free techniques for the most popular HPE methods.

A novel conceptual framework investigating the relationship between roadside advertising and road safety: The driver behaviour and roadside advertising conceptual framework

This paper proposes a conceptual framework to understand the relationship between roadside advertising signs, driver behaviour, and road safety outcomes. Roadside advertising signs are external distractions that may take a driver's attention away from safety-critical driving tasks, potentially increasing crash risk through driver distraction and inattention. Although studies report safety concerns, as a whole, the body of research in the field is inconclusive with inconsistent quality, making it difficult to draw firm conclusions. Definitive links between roadside advertising and road trauma are not yet evident, which has major consequences for road regulators' capacity to develop evidence-based policy to safely administer public roads. However, a lack of consistent evidence does not indicate an absence of risk but underscores its complexity. To address this problem, the Driver Behaviour and Roadside Advertising Conceptual Framework (DBRA framework) was developed to strategically investigate and conceptualise the phenomena of roadside advertising. A new term – “extended engagement” – has also been proposed to account for situations of prolonged attentional engagement with a roadside advertising sign. Further, it is posited that important variations in driving performance may be associated with a driver's extended engagement with a roadside advertising sign. Built on extant theories of driver behaviour and empirical research, the DBRA framework is designed to be a robust tool that encourages a common agenda for future roadside advertising research.

Driver attention prediction based on convolution and transformers

Driver attention prediction based on convolution and transformers

ROAD TRAFFIC SAFETY: A CASE STUDY OF THE PILA POVIAT IN POLAND

Every year, many road accidents occur on Polish roads, in which many people lose their lives and health. Therefore, an attempt was made to analyze road accidents in the Piła poviat. The source of the analysis were the authors' observations and statistical data. This study was conducted on a sample of 250 inhabitants. Hence, a questionnaire was used. Most people, more than half, about 56%, travel by car because it is a more convenient and faster means of transport. While 35% of respondents stated that they feel relatively safe on the roads. A majority, 65%, of the respondents did not participate in a road accident. Further, 60% of the respondents replied that there were many accidents in the Piła poviat, and only 11% answered that there were not many accidents. According to the respondents, the biggest cause of the accidents is alcohol, inattention of drivers and haste. Despite the information, advertisements, scale of these accidents, and opinions of family and friends, people still drive under the influence of alcohol. Year by year, the number of accidents on the roads of the Piła poviat is low, but still considerably high. There is no best way to stop this from happening. In Piła itself, circular intersections have been introduced to improve safety, such an operation is one of the methods for smoother traffic. Additionally, in residential areas, there are more speed bumps to reduce the speed of vehicles. The conclusions obtained in this article can be successfully implemented, not only in other cities in Poland but also in Europe as well.

A Novel Car-Following Model Considering Driver's Attention in Curvy Two-Lane Environment

A Novel Car-Following Model Considering Driver's Attention in Curvy Two-Lane Environment

Design of drowsy and alcohol consumption detection system in vehicles using at mega microcontroller

Driver drowsiness and alcohol consumption are the most common causes of traffic accidents and the resulting financial losses. Researchers have been working on developing driver inattention monitoring systems as drowsy and drunken driving have significantly increased the number of road accidents, which results in serious and deadly collisions. The enormous benefits of sleep detection in road transportation cannot be overstated. The proposed work consists of the design of a drowsy and alcohol consumption detection system in the vehicle using an AT Mega microcontroller monitoring and control system to avoid accidents due to driver inattention. This system consists of an eye blink sensor and an alcohol sensor in the vehicle to alert the driver. This proposed system generates a model which can prevent such accidents.

АНАЛИЗ ПРОБЛЕМ ОЦЕНКИ ПРОФЕССИОНАЛЬНЫХ РИСКОВ ПРИ УВЕЛИЧЕНИИ ДОЛИ ПОЖИЛЫХ ТРАКТОРИСТОВ-МАШИНИСТОВ В СЕЛЬСКОМ ХОЗЯЙСТВЕ

Due to the change in the retirement age and the increase in the proportion of elderly tractor drivers in agricultural enterprises, it became possible to increase the number of emergencies and accidents in the field. The methodological basis of the research was the study and analysis of changes in human cognitive abilities with the increasing age (decreased attention, memory, and performance) and operating environment factors of tractor drivers that affect their fatigue and performance. It has been established that the risks of increased number of emergencies and accidents may grow due to natural biological processes occurring in the body of most people after reaching the age of 60. The age-related decline in attention, memory, decision-making speed, and efficiency of elderly tractor drivers can be aggravated by the presence of harmful production factors in the workplace. At the same time, over many years of work in unfavorable sanitary and hygienic conditions, the risks of occupational diseases may also increase. The author draws conclusions about the need for agricultural managers to adjust the system of both assessing occupational risks and managing the health and safety of elderly tractor drivers. It is advisable for the labor protection services of agricultural enterprises to develop and implement a system for managing the risks of overwork, organizing fatigue preventive measures for elderly tractor drivers and, accordingly, the probability of various incidents and accidents. It is necessary to conduct research on the combined effect of microclimate and noise parameters on the fatigue, reaction speed, and attentiveness of elderly tractor drivers to develop requirements for their work and rest modes and clarify the requirements for the arrangement of tractor cabs and mobile agricultural machines.

Hypo-Driver: A Multiview Driver Fatigue and Distraction Level Detection System

Traffic accidents are caused by driver fatigue or distraction in many cases. To prevent accidents, several low-cost hypovigilance (hypo-V) systems were developed in the past based on a multimodal-hybrid (physiological and behavioral) feature set. Similarly in this paper, real-time driver inattention and fatigue (Hypo-Driver) detection system is proposed through multi-view cameras and biosignal sensors to extract hybrid features. The considered features are derived from non-intrusive sensors that are related to the changes in driving behavior and visual facial expressions. To get enhanced visual facial features in uncontrolled environment, three cameras are deployed on multiview points (0°, 45°, and 90°) of the drivers. To develop a Hypo-Driver system, the physiological signals (electroencephalography (EEG), electrocardiography (ECG), electro-myography (sEMG), and electrooculography (EOG)) and behavioral information (PERCLOS70-80-90%, mouth aspect ratio (MAR), eye aspect ratio (EAR), blinking frequency (BF), head-titled ratio (HT-R)) are collected and pre-processed, then followed by feature selection and fusion techniques. The driver behaviors are classified into five stages such as normal, fatigue, visual inattention, cognitive inattention, and drowsy. This improved hypo-Driver system utilized trained behavioral features by a convolutional neural network (CNNs), recurrent neural network and long short-term memory (RNN-LSTM) model is used to extract physiological features. After fusion of these features, the Hypo-Driver system is classified hypo-V into five stages based on trained layers and dropout-layer in the deep-residual neural network (DRNN) model. To test the performance of a hypo-Driver system, data from 20 drivers are acquired. The results of Hypo-Driver compared to state-of-the-art methods are presented. Compared to the state-of-the-art Hypo-V system, on average, the Hypo-Driver system achieved a detection accuracy (AC) of 96.5%. The obtained results indicate that the Hypo-Driver system based on multimodal and multiview features outperforms other state-of-the-art driver Hypo-V systems by handling many anomalies.

倒立振子型パーソナルトランスポーターの安全性評価に関するシミュレータ構築

In recent years, inverted pendulum personal transporters ( PT ) which have excellent visibility and operability have been used mainly in Europe and the USA, but accidents due to driver inattention have been reported. Therefore, a driving assistance system is needed to help the driver notices obstacles. In the evaluation of driving assistance systems, experiments using simulators are important to ensure safety and to unify conditions. The aim of this study is construction of a simulator and applying it to safety performance evaluation of driving assistance systems. In order to simulate driving in a actual PT, a PT model is constructed using simulator software that moves according to the driver's centre of mass. The view of the PT model is projected. The behaviour of the PT model was compared with the actual PT to evaluate its reproducibility. An evaluation experiment was conducted with a driving assistance device mounted on the control interface. The PT model drives toward the obstacle in the simulator and the driver applies the brakes when the driving assistance device is activated. Drivers were surveyed after simulated driving. Braking distances were compared with speeds.

Drowsiness Detection using EEG signals and Machine Learning Algorithms

Drowsiness is described as a state of reduced consciousness and vigilance accompanied by a desire or want to sleep. Driver tiredness is frequently detected using wearable sensors that track vehicle movement and camera-based systems that track driver behavior. Many alternative EEG-based drowsiness detection systems are developed due to the potential of electroencephalogram (EEG) signals to observe human mood and the ease with which they may be obtained. This paper applies Deep learning architecture like Convolutional Neural networks (CNN) and algorithms for the classification of EEG data for Drowsiness Detection. The key measures of video-based approaches include the detection of physical features; nevertheless, problems such as brightness limitations and practical challenges such as driver attention limits its usefulness. The main measure of video-based methods is the degree of closure of the eyelids; however, its success is limited by constraints like as brightness restrictions and practical challenges such as driver distraction. We have extracted statistical features and trained using various classifiers like Logistic Regression, Naïve Bayes, SVM, and K Nearest Neighbours and compared the accuracy using a deep learning CNN model. Results demonstrate that CNN achieved an accuracy of 94.75% by delegating feature extraction on itself. Upon comparing existing state–of–the–art drowsiness detection systems, the testing results reveal a higher detection capability. The results show that the the suggested method can be used to develop a reliable EEG-based driving drowsiness detection system.

Mtsf: Multi-Scale Temporal-Spatial Fusion Network for Driver Attention Prediction

Mtsf: Multi-Scale Temporal-Spatial Fusion Network for Driver Attention Prediction

The conditioned anticipation of people (CAP) model of driving in urban spaces

Despite significant gains in overall collision rates, pedestrian and bicycle safety in complete street environments remains an on-going challenge. However, urban areas with the most risk exposure for pedestrians continue to maintain lower incident rates than their suburban counterparts (“Dangerous by Design” 2021). Under most driving circumstances, the over-rehearsed nature of driving leads to a psychological state similar to self-hypnosis, where attention is subconsciously maintained on the driving task while metacognitive awareness is minimized or eliminated. This state continues until some type of conflict, uncertainty, and/or novel stimulus is presented. The primary difference in driver attention in urban environments is postulated to result from the Conditioned Anticipation of People (CAP). Based on the human neurological predisposition to recognize and fixate on human faces and figures, we hypothesize that in areas where drivers have been conditioned to expect human presence, low-level metacognition is preemptively re-engaged to address their presence, resulting in higher attentional resource expenditure and increased distraction management. Such conditioning may be generated by contextual features common to pedestrian-friendly environments but, necessarily, must be reinforced over time by the Actual Presence of People (APP). CAP driver engagement is limited by the perceptual abilities of the driver such that at higher speeds or within wider corridors, the presence of pedestrians is more difficult to perceive. This results in a non-CAP attention pattern, exhibiting minimal metacognitive activity, high levels of automaticity, and reduced attention. This model was generated based on the observation that vulnerable user presence and the roadway contextual features that support the driver's ability to see vulnerable users were related to attention data measured during the SHRP2 Naturalistic Driving Study. Visually discernable features that were associated with vulnerable user presence had relationships with attention and large effect sizes (η2 > 0.5). Contextual features that had relationships with vulnerable user presence, but minimal visual impact or interfered with the driver’s ability to see vulnerable users had no relationship to driver attention. This behavioral pattern provides supportive evidence for the proposed model.