Development and classification of autonomous vehicle's ambiguous driving scenario

Human drivers are gradually being replaced by highly automated driving systems, and this trend is expected to persist. The response of autonomous vehicles to Ambiguous Driving Scenarios (ADS) is crucial for legal and safety reasons. Our research focuses on establishing a robust framework for developing ADS in autonomous vehicles and classifying them based on AV user perceptions. To achieve this, we conducted extensive literature reviews, in-depth interviews with industry experts, a comprehensive questionnaire survey, and factor analysis. We created 28 diverse ambiguous driving scenarios and examined 548 AV users' perspectives on moral, ethical, legal, utility, and safety aspects. Based on the results, we grouped ADS, with all of them having the highest user perception of safety. We classified these scenarios where autonomous vehicles yield to others as moral, bottleneck scenarios as ethical, cross-over scenarios as legal, and scenarios where vehicles come to a halt as utility-related. Additionally, this study is expected to make a valuable contribution to the field of self-driving cars by presenting new perspectives on policy and algorithm development, aiming to improve the safety and convenience of autonomous driving.

A Proposed Waiting Time Algorithm for a Prediction and Prevention System of Traffic Accidents Using Smart Sensors

One of the applications of neural networks is to predict the fault section results of traffic utilizing the combined model estimation of the fault section and self-learning models with smart sensors. The prediction of the fault section can autonomously develop the internal model of the network to fit the pre-entered “traffic accident” section data and predict the occurrence of traffic accident sections. In this paper, we propose the results of waiting time for traffic accidents in case of traffic accidents by using a neural network and fuzzy expert system, in comparison with existing algorithms and algorithms for determining traffic accidents. It is used to estimate or predict traffic accident reliability as well. Typically, the type of fault data collected is the number of faults (the number of faults recorded during a given time interval) or the time of fault (the time-of-fault data recorded when each fault occurred), and this can be utilized only for group data types, rather than the time-of-fault data type.

The Validity of Three New Driving Simulator Scenarios: Detecting Differences in Driving Performance by Difficulty and Driver Gender and Age

OBJECTIVE

We explored the convergent and discriminant validity of three driving simulation scenarios by comparing behaviors across gender and age groups, considering what we know about on-road driving.

BACKGROUND

Driving simulators offer a number of benefits, yet their use in real-world driver assessment is rare. More evidence is needed to support their use.

METHOD

A total of 104 participants completed a series of increasingly difficult driving simulation scenarios. Linear mixed models were estimated to determine if behaviors changed with increasing difficulty and whether outcomes varied by age and gender, thereby demonstrating convergent and discriminant validity, respectively.

RESULTS

Drivers adapted velocity, steering, acceleration, and gap acceptance according to difficulty, and the degree of adaptation differed by gender and age for some outcomes. For example, in a construction zone scenario, drivers reduced their mean velocities as congestion increased; males drove an average of 2.30 km/hr faster than females, and older participants drove more slowly than young (5.26 km/hr) and middle-aged drivers (6.59 km/hr). There was also an interaction between age and difficulty; older drivers did not reduce their velocities with increased difficulty.

CONCLUSION

This study provides further support for the ability of driving simulators to elicit behaviors similar to those seen in on-road driving and to differentiate between groups, suggesting that simulators could serve a supportive role in fitness-to-drive evaluations.

APPLICATION

Simulators have the potential to support driver assessment. However, this depends on the development of valid scenarios to benchmark safe driving behavior, and thereby identify deviations from safe driving behavior. The information gained through simulation may supplement other forms of assessment and possibly eliminate the need for on-road testing in some situations.

Effects of Initial Starting Distance and Gap Characteristics on Children's and Young Adults' Velocity Regulation When Intercepting Moving Gaps

OBJECTIVE

This study investigated how children and young adults regulate their velocity when crossing roads under varying traffic conditions.

BACKGROUND

To cross roads safely, pedestrians must adapt their movements to the moving vehicles around them while tightly coupling their movement to visual information.

METHOD

Using an Oculus Rift, 16 children and 16 young adults walked on a treadmill and intercepted gaps between two simulated moving vehicles in an immersive virtual environment. We varied the participants' initial distance from the curb to the interception point, as well as gap characteristics, including gap size and vehicle size.

RESULTS

Varying the initial distance led to systematic adjustments in participants' approach velocities. The inter-vehicle gap and the vehicle size affected the crossing position induced by the initial distance. However, participants did not systematically scale their positions according to the initial distance in narrow gap. Notably, children did not finely tune their movements when they approached wide gap from a closer distance or when they approached the large vehicle from closer distance.

CONCLUSION

Children were less precise in coupling their movements to the moving vehicle in complex traffic environments. In particular, large moving vehicles approaching at closer distances can pose risks when children cross roads.

APPLICATION

These findings suggest the need for an intervention program to improve children's skill in perceiving larger vehicles and timing their movements when crossing roads. We suggest using an interactive virtual reality system to practice this skill.

Susceptibility to social pressure following ventromedial prefrontal cortex damage

Social pressure influences human behavior including risk taking, but the psychological and neural underpinnings of this process are not well understood. We used the human lesion method to probe the role of ventromedial prefrontal cortex (vmPFC) in resisting adverse social pressure in the presence of risk. Thirty-seven participants (11 with vmPFC damage, 12 with brain damage outside the vmPFC and 14 without brain damage) were tested in driving simulator scenarios requiring left-turn decisions across oncoming traffic with varying time gaps between the oncoming vehicles. Social pressure was applied by a virtual driver who honked aggressively from behind. Participants with vmPFC damage were more likely to select smaller and potentially unsafe gaps under social pressure, while gap selection by the comparison groups did not change under social pressure. Participants with vmPFC damage also showed prolonged elevated skin conductance responses (SCR) under social pressure. Comparison groups showed similar initial elevated SCR, which then declined prior to making left-turn decisions. The findings suggest that the vmPFC plays an important role in resisting explicit and immediately present social pressure with potentially negative consequences. The vmPFC appears to contribute to the regulation of emotional responses and the modulation of decision making to optimize long-term outcomes.

Motorcyclist's lane position as a factor in right-of-way violation collisions: a driving simulator study

A driver turning left and failing to notice an oncoming motorcyclist until too late is the most common cause of motorcycle collisions. Consequently, much previous research has focused on motorcycle properties, such as size, shape, and color to explain its inconspicuousness. However, collision statistics remain largely unchanged, suggesting that the issue may not be related solely to the motorcycle's static properties. In the present study, we examined a different characteristic of the motorcycle, namely its trajectory of approach. Seventeen participants faced oncoming traffic in a high-fidelity driving simulator and indicated when gaps were safe enough for them to turn left at an intersection. We manipulated the size of the gaps and the type of oncoming vehicle over 135 trials, with gap sizes varying from 3 to 5s, and vehicles consisting of either a car, a motorcycle in the left-of-lane position, or a motorcycle in the right-of-lane position. Our results show that drivers are more likely to turn in front of an oncoming motorcycle when it travels in the left-of-lane position than when it travels in the right-of-lane position.

Intersection crossing considered as intercepting a moving traffic gap: effects of task and environmental constraints

Safely crossing an intersection requires that drivers actively control their approach to the intersection with respect to characteristics of the flow of incoming traffic. To further our understanding of the perceptual-motor processes involved in this demanding manoeuvre, we designed a driving simulator experiment in which 13 participants actively negotiated intersections by passing through a gap in the train of incoming traffic. Task constraints were manipulated by varying the size of the traffic gap and the initial conditions with respect to the time of arrival of the traffic gap at the intersection. Environment constraints were manipulated by varying the intersection geometry through changes in the angle formed by the crossroads. The results revealed that the task constraints systematically gave rise to continuous and gradual adjustments in approach velocity, initiated well before arriving at the intersection. These functionally appropriate adjustments allowed the drivers to safely cross the intersection, generally just slightly ahead of the center of the traffic gap. Notwithstanding the fact that the geometry of the intersection did not affect the spatiotemporal constraints of the crossing task, approach behavior varied systematically over geometries, suggesting that drivers rely on the traffic gap's bearing angle. Overall, the pattern of results is indicative of a continuous coupling between perception and action, analogous to that observed in locomotor interception tasks.

Effects of major-road vehicle speed and driver age and gender on left-turn gap acceptance

Because the driver's gap-acceptance maneuver is a complex and risky driving behavior, it is a highly concerned topic for traffic safety and operation. Previous studies have mainly focused on the driver's gap acceptance decision itself but did not pay attention to the maneuver process and driving behaviors. Using a driving simulator experiment for left-turn gap acceptance at a stop-controlled intersection, this study evaluated the effects of major traffic speed and driver age and gender on gap acceptance behaviors. The experiment results illustrate relationships among drivers' left-turn gap decision, driver's acceleration rate, steering action, and the influence of the gap-acceptance maneuver on the vehicles in the major traffic stream. The experiment results identified an association between high crash risk and high traffic speed at stop-controlled intersections. The older drivers, especially older female drivers, displayed a conservative driving attitude as a compensation for reduced driving ability, but also showed to be the most vulnerable group for the relatively complex driving maneuvers.

The impact of hands-free message reception/response on driving task performance

A series of closed-course driving experiments were conducted in which 41 drivers ranging in age from 19 to 70 were put through a series of increasingly challenging driving performance tasks both in the presence and absence of audible messages. The messages required specific responses and these, along with driving performance measures based on driver/vehicle response characteristics, were recorded. The results clearly showed a negative impact of the message task on driver decision-making performance when this involved the more complex tasks of weaving and especially left-turning. Such decision-making decrements in the presence of the messages were exacerbated by adverse pavement surface conditions.

Factors influencing the probability of an incident at a junction: results from an interactive driving simulator

Using data generated from a fixed-base interactive driving simulator, which was used to evaluate a driver decision aid, a model is built to predict the probability of an incident (i.e. an accident or a 'near miss') occurring as a result of a right-turn across left-hand traffic at an unsignalised junction. This can be considered to be the product of two separate probabilities, the first being the probability that the gap between a pair of vehicles in the traffic stream is accepted, and the second the probability that the time needed to cross the on-coming stream of traffic causes the time-to-collision with the nearest vehicle in this traffic stream to be less than a second. The model is developed from the results of experimental trials involving a sample of drivers, the majority of whom were aged 60 years or older, in order to demonstrate the effect of various parameters on these probabilities. The parameters considered include the size of the gap between successive vehicles, vehicle characteristics such as size, colour and velocity, driver characteristics such as age and sex, and both daytime and night-time conditions.

Social psychological models of choice behavior and drivers' left turns

ABSTRACT. The authors reanalyzed data from a simulated left-turn experiment (P. A. Hancock, J. K. Caird, S. Shekhar, & M. Vercruyssen, 1991) to test the adequacy of the nonlinear Gray-Tallman satisfaction balance model of choice behavior (L. N. Gray & I. Tallman, 1984) in predicting left turns. Participants (Hancock et al., 1991) were 40 experienced U.S. drivers who were exposed to simulated oncoming traffic; the size of the vehicle (motorcycle, compact vehicle, full-sized vehicle, delivery truck), its speed (10-70 mph, or 16-112 kmph), and the intervehicle time gap (3-9 s) varied. Hancock et al. (1991) measured (a) the likelihood of a left turn and (b) the occurrence of a collision. The probability of a left turn was greater for larger intervehicle time gaps and for oncoming smaller vehicles traveling at higher speeds. The Gray-Tallman (1984) model explained 69% of the variation in turning versus 57% for a linear regression model. In making decisions people tend to treat the values and costs affecting choices in a multiplicative, rather than linear, fashion. The Gray-Tallman model also has the potential for incorporating, both theoretically and mathematically, an unlimited range of potential values and costs that may influence left turn decisions.