Detection of Driver Distraction Using Vision-Based Algorithms (original) (raw)
Related papers
Transportation Research Part F-traffic Psychology and Behaviour, 2005
The relationship between distraction and driving performance: towards a test regime for in-vehicle information systems Commonly special issues are prepared on a specific topic, either on the basis of invited papers to cover a certain research area, or they are based on selected papers from a conference. However, occasionally the common accomplished mission and coherence of a multi-partner project warrant a special issue as well. The now completed European project HASTE (Human machine interface And the Safety of Traffic in Europe) aimed to develop methodologies and guidelines for the assessment of In-Vehicle Information Systems (IVIS). To date, there have been several attempts to provide manufacturers and testing authorities with a set of guidelines to assess the likely impacts of IVIS on the driving task, usually in the form of a checklist. Such checklists provide a tool that enables the identification of likely problems but they do not enable the quantification of safety problems. This project is fundamental to the development of a valid, reliable and efficient tool that will aid testing authorities in their safety evaluation of IVIS. The raison dÕêtre of the effort is quite straightforward. Lately, a large number of technological developments have enabled the rapid increase of applications that can be applied for delivering in the vehicle traffic information and other forms of driver support. In the past the main targeted group was professional drivers but more recently there has been an explosion in systems catering to drivers of private cars. Advanced traffic information and support systems and services potentially contribute to a safe, efficient and environmentally friendly traffic flow. Accurate and timely traffic information can decrease travel times and costs, and also momentary workload as the driver does not have to divert his/her attention to other sources of travel information. However, a potential negative side effect from certain aspects of these new methods of traffic information provision arises from the growing amount of information addressed at road-users. Every new information source could add to the information load of drivers, potentially counteracting the potential benefits of decreased workload from the same information. In order to control negative traffic safety effects, it is necessary to adapt traffic information presentation to current traffic situations and road-user requirements. Only if traffic information is easily accessible, carefully timed, understandable and matched to user needs, will information overload on drivers be
Accident Analysis & Prevention, 2007
In this on-road experiment, drivers performed demanding cognitive tasks while driving in city traffic. All task interactions were carried out in hands-free mode so that the 21 drivers were not required to take their visual attention away from the road or to manually interact with a device inside the vehicle. Visual behavior and vehicle control were assessed while they drove an 8 km city route under three conditions: no additional task, easy cognitive task and difficult cognitive task. Changes in visual behavior were most apparent when performance between the No Task and Difficult Task conditions were compared. When looking outside of the vehicle, drivers spent more time looking centrally ahead and spent less time looking to the areas in the periphery. Drivers also reduced their visual monitoring of the instruments and mirrors, with some drivers abandoning these tasks entirely. When approaching and driving through intersections, drivers made fewer inspection glances to traffic lights compared to the No Task condition and their scanning of intersection areas to the right was also reduced. Vehicle control was also affected; during the most difficult cognitive tasks there were more occurrences of hard braking. Although hands-free designs for telematics devices are intended to reduce or eliminate the distraction arising from manual operation of these units, the potential for cognitive distraction associated with their use must also be considered and appropriately assessed. These changes are captured in measures of drivers' visual behavior.
Real-time evaluation of driver’s alertness on highways
Urban Transport XVII, 2011
Suburbanisation has been internationally a major phenomenon in the last decades. Suburb-to-suburb routes are nowadays the most widespread road journeys; and this resulted in an increment of distances travelled, particularly on faster suburban highways. The design of highways tends to over-simplify the driving task and this can result in decreased alertness. Driving behaviour is consequently impaired and drivers are then more likely to be involved in road crashes. This is particularly dangerous on highways where the speed limit is high. While effective countermeasures to this decrement in alertness do not currently exist, the development of in-vehicle sensors opens avenues for monitoring driving behaviour in real-time. The aim of this study is to evaluate in real-time the level of alertness of the driver through surrogate measures that can be collected from in-vehicle sensors. Slow EEG activity is used as a reference to evaluate driver's alertness. Data are collected in a driving simulator instrumented with an eye tracking system, a heart rate monitor and an electrodermal activity device (N=25 participants). Four different types of highways (driving scenario of 40 minutes each) are implemented through the variation of the road design (amount of curves and hills) and the roadside environment (amount of buildings and traffic). We show with Neural Networks that reduced alertness can be detected in real-time with an accuracy of 92% using lane positioning, steering wheel movement, head rotation, blink frequency, heart rate variability and skin conductance level. Such results show that it is possible to assess driver's alertness with surrogate measures. Such methodology could be used to warn drivers of their alertness level through the development of an in-vehicle device monitoring in realtime drivers' behaviour on highways, and therefore it could result in improved road safety.
First-Stage Evaluation of a Prototype Driver Distraction Human-Machine-Interface Warning System
Journal of Road Safety, 2021
Recent advances in vehicle technology permit the real-time monitoring of driver state to reduce distraction-related crashes, particularly within the heavy vehicle industry. Relatively little published research has evaluated the human machine interface (HMI) design for these systems. However, the efficacy of in-vehicle technology depends in large part on the acceptability among drivers of the system’s interface. Four variations of the HMI of a prototype multi-modal warning system developed by the authors for driver distraction were evaluated in a truck simulator with eight car drivers and six truck drivers. Driver acceptance of the HMIs was assessed using the System Acceptability Scale; and salience, comprehension and perceived effectiveness of components of the HMIs (modality, intensity of warning) were assessed using likert scales. The results showed that participants considered the HMIs to be acceptable and useful, and that the warning components were largely noticed, understood c...
Validation of in-car observations, a method for driver assessment
Transportation Research Part A: Policy and Practice, 2004
An in-car observation method with human observers in the car was studied to establish whether observers could be trained to observe safety variables and register driverÕs behaviour in a correct and coherent way, and whether the drivers drove in their normal driving style, despite the presence of the observers. The study further discussed the observed variables from a safety perspective. First three observers were trained in the observation method and on-road observations were carried out. Their observations were then compared with a key representing a correct observation. After practising the observation method the observers showed a high correlation with the key. To establish whether the test drivers drove in a normal way during the in-car observations, comparisons of 238 spot-speed measurements were carried out. DriverÕs speeds when driving their own private cars were compared with their speeds during the in-car observations. The analysis showed that the drivers drove in the same way when being observed as they did normally. Most of the variables studied in the in-car observations had a well documented relevance to traffic safety. Overall, in-car observation was shown to be a reliable and valid method to observe driver behaviour, and observed changes provide relevant data on traffic safety. drivers and by accident statistics Highway Loss Data Institute, 1994). Typically, driver behaviour is studied using driving simulators, instrumented vehicles or human observers inside the vehicle. While all these methods have their advantages,their disadvantages are that they can generate data that is not always reliable or valid.
IEEE Transactions on Intelligent Transportation Systems, 2013
Driver distraction detection and intervention are important for designing modern driver-assistance systems and for improving safety. The main research question of this paper is to investigate how the cumulative driver off-road glance duration can be controlled to reduce the probability of occurrences of crash and near-crash events. Based on the available data sets from the Virginia Tech Transportation Institute (VTTI) 100-car study, the conditional probability is calculated to study the chance of crash and near-crash events when the given cumulative off-road glance duration in 6 s has been reached. Different off-road eye-glance locations and traffic density levels are also evaluated. The results show that one linear relationship can be obtained between the cumulative off-road eye-glance duration in 6 s and the risk of occurrences of crash and near-crash events, which varies for different off-road eye-glance locations. In addition, the traffic density level is found to be one significant moderator to this linear relationship. Detailed comparisons are made for different traffic density levels, and one nonlinear equation is obtained to predict the probability of occurrences of crash and near-crash events by considering both cumulative off-road glance duration and traffic density levels.
Visual in-car warnings: How fast do drivers respond?
Transportation Research Part F: Traffic Psychology and Behaviour, 2018
We investigate how quickly drivers can change lanes in response to a visual in-car warning. Our work is motivated by technological developments, in which beacons along the road can trigger in-car warnings, for example when a driver is approaching a lane closure. What is not known, however, is at what distance such an in-car warning still allows for a timely lane change. We measured how quickly drivers respond to a visual in-car warning in a driving simulator. The driving task was combined with an audio task that provided different levels of cognitive distraction. We found that the initial reaction time to in-car warnings was significantly larger for drivers that were distracted by the audio task. Although the majority of drivers responded in time for a safe lane change, some drivers occasionally missed these signals, pointing at a serious potential hazard. Indeed, the results of a simulation model, used to investigate how this might extrapolate to regular traffic conditions, suggest that around 50% of drivers might not make a timely lane change in response to a last-minute warning. This indicates that these signals might be insufficient on their own when applied in the real world. This work can inform the design and evaluation of safer roads and in-car interfaces.
2011
−There has been recent interest in intelligent vehicle technologies, such as advanced driver assistance systems (ADASs) or in-vehicle information systems (IVISs), that offer a significant enhancement of safety and convenience to drivers and passengers. However, the use of ADASand IVIS-based information devices may increase driver distraction and workload, which in turn can increase the chance of traffic accidents. The number of traffic accidents involving older drivers that are due to distraction, misjudgment, and delayed detection of danger, all of which are related to the drivers' declining physical and cognitive capabilities, has increased. Because the death rate in traffic accidents is higher when older drivers are involved, finding ways to reduce the distraction and workload of older drivers is important. This paper generalizes driver information device operations and assesses the workload while driving by means of experiments involving 40 drivers in real cars under actual ...
Driving performance assessment: Effects of traffic accident location and alarm content
Accident Analysis & Prevention, 2008
According to accident statistics for Taiwan, the two most common traffic accident locations in urban areas are roadway segments and intersections. On roadway segments, most collisions are due to drivers not noticing the status of leading vehicle. At intersections, most collisions are due to the other driver failing to obey traffic signs. Using a driving simulator equipped with a collision warning system, this study investigated driving performance at different accident locations and between different alarm contents, and identified the relationship between crash occurrences and driving performance. Thirty participants, aged 20-29 years, were recruited in this study. Driving performance measures were perception-reaction time, movement-reaction time, speed and a crash. Experimental results indicated that due to different demands for processing information under different traffic conditions, driving performance differed at the two traffic accident locations. On a roadway segment, perception-reaction time for a beep was shorter than the time for a speech message. Nevertheless, at an intersection, a speech message was a great help to drivers and, thus, perception-reaction time was effectively reduced. In addition, logistic regression analysis indicates that perception-movement time had the greatest influence on crash occurrence.