Review of the application of neural network approaches in pedestrian dynamics studies

In recent years, artificial intelligence methods have been widely used in the study of pedestrian dynamics and crowd evacuation. Different neural network models have been proposed and tested using publicly available pedestrian datasets. These studies have shown that different neural network models present large performance differences for different crowd scenarios. To help future research select more appropriate models, this article presents a review of the application of neural network methods in pedestrian dynamics studies. The studies are classified into two categories: pedestrian trajectory prediction and pedestrian behavior prediction. Both categories are discussed in detail from a conceptual perspective, as well as from the viewpoints of methodology, measurement, and results. The review found that the mainstream method of pedestrian trajectory prediction is currently the LSTM-based method, which has adequate accuracy for short-term predictions. Furthermore, the deep neural network is the most popular method for pedestrian behavior prediction. This method can emulate the decision-making process in a complex environment, and it has the potential to revolutionize the study of pedestrian dynamics. Overall, it is found that new methods and datasets are still required to systemize the study of pedestrian dynamics and eventually ensure its wide-scale application in industry.

An observational study on pedestrian and bicyclist violations at railroad grade crossings: Exploring the impact of geometrical and operational attributes

INTRODUCTION

In order to enhance the existing body of research regarding the prevalence of violations committed by pedestrians and bicyclists, this observational study utilized personal-specific, train-specific, operational, and behavioral data extracted from recorded videos at seven railroad grade crossings (RRGCs).

METHOD

A total of 7,332 pedestrians and 1,684 bicyclist crossing events were observed during 279 gate operation phases at seven study locations. First, two separate violation models (binary logit) for pedestrians and bicyclists were developed to determine the influence of personal-specific characteristics and other behavioral traits on probability of pedestrians' and bicyclists' violation. Later, the impact of various geometrical (railway track, distance between gates, distance between gate and nearest track, and road width) and operational (train vehicle unit) parameters on the factors that caused non-motorist violations was examined by developing distinctbinary logit models.

RESULTS

The probability of pedestrians'and bicyclists' violationwas found to be significantly influenced by train characteristics (train speed and train occupancy time), gate operational attributes (warning time and gate blockage duration), and personal-specificcharacteristics (gender, age, crossing speed, and looking both side). Furthermore, geometrical attributes along with train vehicle unit were observed to have impact over various factors that significantly influenced the behavior of pedestrians and bicyclists to violate.

PRACTICAL APPLICATIONS

The findings of this research provide essential information that could be applied in the initiatives of user awareness and enforcement program, emphasis on situational awareness and design of operational and geometricalelementsto significantlylimitthe effects of risky walking andbicycling behavior at RRGCs.

The 'invisible gorilla' during pedestrian-AV interaction: Effects of secondary tasks on pedestrians' reaction to eHMIs

In road traffic, mental overload often leads to a failure to notice new and distinctive stimuli. Such phenomenon is known as 'inattentional blindness'. Safe and efficient interaction between automated vehicles (AVs) and pedestrians is expected to rely heavily on external human-machine interfaces (eHMIs), a tool AVs are equipped with to communicate their intentions to pedestrians. This study seeks to explore the phenomenon of 'inattentional blindness' in the context of pedestrian-AV interactions. Specifically, the aim is to understand the effects of a warning eHMI on pedestrians' crossing decisions when they are engaged in a secondary task. In an experiment study with videos of pedestrian crossing scenarios filmed from the perspective of the crossing pedestrian, participants had to decide the latest point at which they would be willing to cross the road in front of an AV with an eHMI vs. an AV without an eHMI. Participants were also asked to predict the future behavior of the AV. 125 female and 9 male participants aged between 18 and 25 completed the experiment and a follow-up questionnaire. It was found that the presence of a warning eHMI on AVs contributes to a clearer understanding of pedestrians' inferences about the intention of AVs and helps deter late and dangerous crossing decisions made by pedestrians. However, the eHMI fail to help pedestrians avoid such decisions when they face a high mental workload induced by secondary task engagement.

Analysis of walking speeds and success rates on mid-block crossings using virtual reality simulation

A focus is set worldwide to study pedestrian behavior in road situations associated with a high frequency of crashes. This paper presents the results from a virtual reality simulation study that recorded pedestrians performing mid-block crossings on a straight segment of an urban street. The experiment was designed with eight scenarios with combinations of one or two lanes, two vehicle speeds, and constant or variable gaps between vehicles. The experiment was conducted with 48 subjects producing 538 crossing observations. The results show that subjects, on average, watched about 5 vehicle gaps in traffic before crossing the street and accepted a gap of 4.5 s between vehicles to cross. A regression model showed that the vehicle speed, the number of lanes, and subjects in the 66-85 years old group had a significant effect on the gap value accepted to cross. An interaction term based on gender and number of lanes also had a significant effect on the accepted gap. The study found average walking speeds between 4.1 and 4.8 ft/s (1.2-1.4 m/s) for different scenarios. A regression model revealed that the gender, the number of lanes, and the gap accepted to cross influenced the walking speed. Significant effects of interactions of the age with gender, number of lanes, and vehicle speed variables were also found that explain the differences in walking speed. The results for the success rate when crossing the street showed the overall worst performance in the scenario with traffic generated with a 25-mph (40 km/h) speed and a constant 3-s gap between vehicles. A Logit model showed that the probability of a pedestrian being hit by a vehicle increased with age, with traffic at the top vehicle speed, and with the constant 3-s vehicle gap. In contrast, the probability decreased with increases in the vehicle gap accepted to cross and the walking speed.

Pedestrian safety at roundabouts: Their crossing and glance behavior in the interaction with vehicular traffic

Smartphones are nowadays indispensable devices in daily life. Their rapid technological development makes it possible to have almost all necessary information on them, which facilitates their spread among all users. However, this extensive use has led to many distraction problems while performing other main tasks, such as activities on the road, which can also impact people's safety. Therefore, various experts focused their attention on issues related to drivers looking at their mobile phones; recently, researchers from different disciplines saw the need to deepen knowledge also on the phenomenon of vulnerable road users distracted by digital devices, especially when approaching signalized intersections. This study builds on and seeks to extend this area of research by analyzing the effects of digital distraction on pedestrians as they approach unsignalised intersections located on roundabout entrances and exits. The aim of the research is to understand the extent to which the task of checking social media affects pedestrian reaction and crossing times, as well as to identify which elements attract the most pedestrian attention. To achieve this goal, an eye-tracking study was designed, in which participants wearing eye-tracking glasses were asked to walk a predefined route, once checking their social media apps and once walking without distracting technological elements. The results showed an 84% increase in reaction time when using the phone, while only a slight rise in crossing time was found. The general conclusions about the most observed elements when walking are also consistent with the main findings of previous literature studies.

Factors affecting pedestrian behaviors at signalized crosswalks: An empirical study

INTRODUCTION

Safety of pedestrians depends, among other factors, on their behavior while crossing the road. This study aims to assess behaviors of pedestrians at signalized crosswalks.

METHOD

Following a literature review and a pilot study, 25 vital pedestrian crossing factors and behaviors were determined. Then data was randomly collected for 708 pedestrians at 10 lighted crossings in Sharjah (UAE), five at road intersections and five mid-block crossings.

RESULTS

Results indicated that 17.4% of pedestrians observed crossed partly or fully on red and that crossing speed was 1.22 m/s, on the average, which is slightly faster than most speeds recorded in the literature. Moreover, female pedestrians were more likely to cross while chatting with others, less likely to cross on red, and more likely to walk slower than male pedestrians. Results also showed that pedestrians who crossed at road intersections walked slower than those who crossed at mid-block crossings. It was also found that longer red pedestrian times and narrower roads tended to encourage pedestrians to cross on red and that the majority of pedestrians did not look around before crossing.

PRACTICAL IMPLICATIONS

Use of the Health Belief Model for pedestrian safety are discussed.

The paradox of pedestrian's risk aversion

Traffic accidents are becoming a significant cause for unnatural deaths around the world, with more than 1.25 million fatalities in road accidents each year, and over 20 million people severely injured. A large portion of accidents that result in fatalities involve interaction between vehicles and pedestrians. In the literature, researchers speculate on a wide range of reasons for these figures. This paper focuses on the relationship between pedestrians' urgency to cross a busy road and the resulting level of risk for an accident. The probability for an accident is determined by a prediction model for a collision between drivers and pedestrians at congested conflict spots. The model is based on a motion planner called the Probabilistic Navigation Function (PNF), initially designed for robot navigation in dynamic cluttered and uncertain environments. The model predicts pedestrians' trajectories when crossing a busy road in a sub-meter accuracy, based on the risk they are willing to take (a reflection of the level of urgency to cross the road). The paper describes an unexpected and surprising pedestrian behavior in simple road crossings scenarios. When the model is given a loose risk boundary (that reflects a high level of pedestrian urgency to cross), the resulting trajectory exposes the pedestrian to a lower risk compared with a trajectory constructed with a strict risk boundary (that reflects a more conservative pedestrian). This is equivalent to claiming that, paradoxically, pedestrians in some scenarios who are willing to take higher levels of risk, face a decreased probability for an accident while crossing a congested road. The paper introduces the PNF model for crossing pedestrians, analyses their performance in a set of simulations, and explains its rationale. Next, an analytic estimation for the risk level as a function of the crossing angle of the selected trajectory is provided. A series of experiments conclude the paper and support the claim that this phenomenon is frequent among crossing pedestrians. The experimental results suggest that in some common scenarios, more cautious pedestrians may lower the initial risk for an accident at the expense of a total higher risk for an accident during the entire road crossing process, compared with a pedestrian who takes an initial higher level of risk that results in, overall, a decreased probability for an accident. A statistical analysis implies that there are significant differences in this occurrence between adults and children.

How do fatalistic beliefs affect the attitudes and pedestrian behaviours of road users in different countries? A cross-cultural study

This paper reports on an exploratory investigation of the influence of five different fatalistic belief constructs (divine control, luck, helplessness, internality, and general fatalism) on three classes of self-reported pedestrian behaviours (memory and attention errors, rule violations, and aggressive behaviours) and on respondents' general attitudes to road safety, and how relationships between constructs differ across countries. A survey of over 3400 respondents across Bangladesh, China, Kenya, Thailand, the UK, and Vietnam revealed a similar pattern for most of the relationships assessed, in most countries; those who reported higher fatalistic beliefs or more external attributions of causality also reported performing riskier pedestrian behaviours and holding more dangerous attitudes to road safety. The strengths of relationships between constructs did, however, differ by country, behaviour type, and aspect of fatalism. One particularly notable country difference was that in Bangladesh and, to a lesser extent, in Kenya, a stronger belief in divine influence over one's life was associated with safer attitudes and behaviours, whereas where significant relationships existed in the other countries the opposite was true. In some cases, the effect of fatalistic beliefs on self-reported behaviours was mediated through attitudes, in other cases the effect was direct. Results are discussed in terms of the need to consider the effect of locus of control and attributions of causality on attitudes and behaviours, and the need to understand the differences between countries therein.

Distracted Pedestrian Behavior on two Urban College Campuses

Pedestrian injuries injure about 180,000 individuals and kill 6000 each year in the United States, and pedestrian injury rates have increased each of the last several years. Distracted pedestrian behavior may play a role in the trend of increasing risk for pedestrian injury. Using in vivo behavioral coding over the course of  two weeks on two urban college campuses, this study aimed to (1) understand the type and rate of distractions engaged in by pedestrians on urban college campuses, and (2) investigate the impact of distraction on street-crossing safety and behavior. A total of 10,543 pedestrians were observed, 90% of them young adults. Over one-third of those pedestrians were distracted while actively crossing roadways. Headphones were the most common distraction (19% of all pedestrians), followed by text-messaging (8%) and talking on the phone (5%). Women were more likely to text and talk on the phone than men, and men were more likely to be wearing headphones. Distracted pedestrians were somewhat less likely to look for traffic when they entered roadways. As handheld device usage continues to increase, behavioral interventions should be developed and implemented. Changes to policy concerning distracted pedestrian behavior, including improvement of the built environment to reduce pedestrian risk, should be considered in busy pedestrian areas like urban college campuses.

Gender differences in children's pedestrian behaviors: Developmental effects

INTRODUCTION

All child pedestrians are vulnerable to road traffic injuries, but there is evidence that boys may be at greater risk than girls, at least among some age groups.

METHOD

To create effective intervention programs, research on how boys and girls of particular ages behave as pedestrians, and whether there are gender differences that increase risk for particular genders at particular ages, is needed. In this study, 255 boys and 235 girls in grade 1 through 6 from two primary schools in Nantong city, China, were unobtrusively videotaped when walking to school in the morning. Videotapes were reviewed and coded for seven pedestrian safety behaviors, including activities related to crossing streets (walking instead of running/hopping, observing traffic, using the crosswalk instead of walking outside the crosswalk, and keeping close to an accompanying adult) and those related to walking alongside busy streets (walking instead of running/hopping on the sidewalk, not playing on the sidewalk, walking alone instead of walking side-by-side with a partner).

RESULTS

Results revealed that as a whole, boys played on the sidewalk more often than girls (p < 0.01) and crossed with an accompanying adult more than girls (p < 0.05), while girls walked side-by-side with partners more often than boys (p < 0.05). With a few exceptions, boys and girls in the younger grades (1-2) as well as those in the older grades (5-6) behaved fairly similarly as pedestrians, but boys and girls in the middle grades (3-4) presented with several significant gender-based differences. In the middle grades, boys watched traffic more than girls while crossing (p < 0.01); ran, hopped and played on the sidewalk more often than girls (p < 0.05); and walked side-by-side less often with partners than girls (p < 0.05). We also detected different gender-based trends in the development of pedestrian skills. With increasing age, girls performed more safely in pedestrian tasks, but boys did not show a similar developmental trend.

CONCLUSIONS

We conclude that boys and girls exhibit different characteristics in their pedestrian behaviors and discuss implications for prevention.

The mediating effect of traffic safety climate between pedestrian inconvenience and pedestrian behavior

Because most people are pedestrians at some point on any given day and walking is the most indispensable means of transportation, pedestrian safety should be investigated. The primary aim of this study was to investigate the relationships among the inconveniences that pedestrians perceive in city traffic, the traffic safety climate and pedestrian behavior. A total of 311 participants voluntarily and validly completed a survey that included the Pedestrian Inconvenience Questionnaire (PIQ), the Traffic Climate Scale (TCS) and the Pedestrian Behavior Scale (PBS). We discovered that pedestrians' perceived inconvenience was positively correlated with transgression and positive behavior by pedestrians and it also positively correlated with the external affective demands (emotional engagement facet of TCS) while negatively correlated with the functionality (functional traffic system facet of TCS). We determined that the external affective demands were positively correlated with pedestrian risk behaviors (i.e., transgression, aggressive behaviors and lapses), internal requirements (traffic participants' skills facet of TCS) were positively correlated with positive behaviors, and functionality was negatively correlated with transgression and lapses. Moreover, the results indicate that the relationship between the inconveniences pedestrians perceive in city traffic and pedestrians' transgressive behavior was fully mediated by the functionality dimension of the traffic safety climate. Pedestrians' perceived inconvenience is an important factor that affects pedestrian behavior, and the influence of pedestrians' perceptions of the traffic safety climate cannot be disregarded.

Injury severity analysis in taxi-pedestrian crashes: An application of reconstructed crash data using a vehicle black box

In-vehicle recording devices have enabled recent changes in methodological paradigms for traffic safety research. Such devices include event data recorders (EDRs), vehicle black boxes (VBBs), and various sensors used in naturalistic driving studies (NDSs). These technologies may help improve the validity of models used to assess impacts on traffic safety. The objective of this study is to analyze the injury severity in taxi-pedestrian crashes using the accurate crash data from VBBs, such as the time-to-collision (TTC), speed, angle, and region of the crash. VBB data from a two-year period (2010-2011) were collected from taxis operating in Incheon, South Korea. An ordered probit model was then applied to analyze the injury severity in crashes. Five variables were found to have a greater effect on injury severity: crash speed, crashes in no-median sections, crashes where the secondary impact object of pedestrians was the crash vehicle, crashes where the third impact object of pedestrians was another moving vehicle, and crashes where the third impact region of pedestrians was their head. However, injuries were less severe in crashes where the first impact region on the pedestrian was their leg, crashes with the car moving in a straight line, and crashes involving junior high school students.

Reducing the impact of speed dispersion on subway corridor flow

The rapid increase in the volume of subway passengers in Beijing has necessitated higher requirements for the safety and efficiency of subway corridors. Speed dispersion is an important factor that affects safety and efficiency. This paper aims to analyze the management control methods for reducing pedestrian speed dispersion in subways. The characteristics of the speed dispersion of pedestrian flow were analyzed according to field videos. The control measurements which were conducted by placing traffic signs, yellow marking, and guardrail were proposed to alleviate speed dispersion. The results showed that the methods of placing traffic signs, yellow marking, and a guardrail improved safety and efficiency for all four volumes of pedestrian traffic flow, and the best-performing control measurement was guardrails. Furthermore, guardrails' optimal position and design measurements were explored. The research findings provide a rationale for subway managers in optimizing pedestrian traffic flow in subway corridors.