No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Drivers’ rear end collision avoidance behaviors under different levels of situational urgency
... Their results showed that the braking response time was longer when the deceleration rate was smaller; conversely, the greater the deceleration rate of the leading vehicle, the shorter the reaction time of the following driver. Wang et al. [15] investigated the relationship between reaction time and driver braking behavior under different levels of urgency, three rates of deceleration (0.3, 0.5, and 0.75 g), and two different headways (1.5 and 2.5 s). Their results showed that shorter distances could produce faster reaction times, and higher deceleration rates could lead to faster reaction times than lower rates. ...
... With regard to deceleration rate, previous studies have indicated that different deceleration rates affect the braking response time of following drivers [15,19]. Wood and Zhang [27] reported that the minimum and maximum deceleration rates among 2971 natural drivers were 0.23 g and 1.09 g, respectively. ...
... where g = 9.8 m/s 2 and µ = 0.8 (coefficient of friction). With regard to deceleration rate, previous studies have indicated that different deceleration rates affect the braking response time of following drivers [15,19]. Wood and Zhang [27] reported that the minimum and maximum deceleration rates among 2971 natural drivers were 0.23 g and 1.09 g, respectively. ...
To prevent vehicle crashes, studies have proposed the use of flashing signals (brake lights or other light indicators) to improve the driver’s response time when the leading vehicle is braking. However, there are no consistent results on the ideal flashing frequency of the brake lights. This study aimed to investigate different brake light flashing frequencies to assess their impact on braking response time. Twenty-four participants aged 25 to 30 were recruited. Two driving speed environments (50 and 80 km/h), three deceleration rates (0.25, 0.6, and 1 g), and four brake light flashing frequencies (0, 2, 4, and 7 Hz) were examined. Braking response time, average braking force, and braking response time ratio were used to evaluate the driving behavior. The results showed that the braking response time and average braking force were affected by the deceleration rate in the 50 km/h driving environment. In the 50 and 80 km/h driving environments, although there were no significant differences among the three deceleration rates, the braking response time decreased by 3–7% under the flashing brake light condition. These findings can be used as a reference for safety designs as well as future studies on driving behavior.
... Drivers' reaction time to a rear-end collision is one of the key parameters that may change according to situations (Aust et al., 2013;Li et al., 2016;Liebermann et al., 1995;Wang et al., 2016;Xue et al., 2018). Liebermann et al. (1995) reported that drivers reacted faster when the following distance got shorter. ...
... Hulst and Der (1999) further found that drivers could adjust their reaction time to the deceleration level of the leading vehicle. Wang et al. (2016) tested the effect of situational urgency on driver's rear-end collision avoidance behaviors and reported that an increased situational urgency could lead to a faster deceleration. Given the varied driver response in different situations, the collision avoidance systems should take driver's real-time response into account when evaluating the situation and making the decisions of warning or autonomous control. ...
... Different from the warning system, the autonomous collision avoidance system, as the last defensive line, needed to interfere with driver's behavior when the danger came, and thus effectively reduce the crash rates (Cicchino, 2017). In the previous studies of autonomous braking system's algorithm, researchers usually used the speed and position information of the leading vehicle to infer the safe distance (Bella & Russo, 2011;Doi, 1994;Wang et al., 2016). However, the changes in driver behavior were often unpredictable, and the variation of deceleration and reaction time would also reduce the accuracy of the autonomous emergency braking system (Xiong et al., 2019). ...
Introduction:
To assist drivers in avoiding rear-end collisions, many early warning systems have been developed up to date. Autonomous braking technology is also used as the last defense to ensure driver's safety.
Method:
By taking the accuracy and timeliness of automatic system control into account, this paper proposes a rear-end Real-Time Autonomous Emergency Braking (RTAEB) system. The system inserts brake intervention based on drivers' real-time conflict identification and collision avoidance performance. A driving simulator-based experiment under different traffic conditions and deceleration scenarios were conducted to test the different thresholds to trigger intervention and the intervention outcomes. The system effectiveness is verified by four evaluation indexes, including collision avoidance rate, accuracy rate, sensitivity rate, and precision rate.
Results:
The results showed that the system could help avoid all collision events successfully and enlarge the final headway distance, and a TTC threshold of 1.5 s and a maximum deceleration threshold of -7.5 m/s2 could achieve the best collision avoidance effect. The paper demonstrates the situations that are more inclined to trigger the RTAEB (i.e., a sudden brake of the leading vehicle and a small car-following distance). Moreover, the study shows that driver characteristics (i.e., gender and profession) have no significant association with system trigger. Practical Applications: The study suggests that development of collision avoidance systems design should pay attention to both the real-time traffic situation and drivers' collision avoidance capability under the present situation.
... The dominant analytical approach to safety thresholds is based on pre-designed kinematic models, such as the collision avoidance safety warning distance model [14][15][16][17][18]. It involves specific metrics like TTC and driver reaction time in different conditions. ...
... It involves specific metrics like TTC and driver reaction time in different conditions. Based on data statistics of dangerous driving scenarios, rear-end collisions account for about 30% of all collision accidents [14], and emergency braking of the preceding vehicle is the main cause of such accidents. Therefore, most researchers focus on the high-risk scenario of emergency braking. ...
Traffic regulations provide a solid foundation for the safety of all road users; however, the ambiguous provisions and unclear safety thresholds within these regulations pose significant challenges to compliance, particularly concerning the safe operation of autonomous vehicles. To address this issue, this paper conducts an in-depth analysis of vehicle emergency braking behavior based on the Aerial Dataset for China Congested Highway and Expressway (AD4CHE). The extraction method for the emergency braking risk scenario of natural driving data is proposed, and the correlation between safe distance, safe speed, and driving safety under the scenario of a slightly congested expressway is elaborated in detail. The safety threshold of ambiguous traffic rules obtained can be used for the digitalization of traffic rules that can support the functional development and traffic safety testing of automated driving systems.
... The work presented in this paper addresses only the first step of this crash scenario generation approach. Many other studies have analyzed the following vehicle's behavior during rear-end emergencies (crashes and near-crashes) by means of a driver response model [18], [19], [20], [21], [22], [23], [24], [25]. For example, Markkula et al. [23] used a piecewise linear model and used driver glance behaviors to model the following vehicles' speed profiles in naturalistic rear-end emergencies. ...
... However, there has been a notable lack of research on the lead vehicle's behavior in such situations despite its significant influence on the following vehicle. In crash reconstruction and rear-end emergency studies, it is commonly assumed that the lead vehicle maintains a constant acceleration or deceleration before the crash [19], [21], [24], even though there is inadequate evidence to support this assumption. To address this knowledge gap, the objective of this study is to develop a model of the lead-vehicle kinematics in rear-end crashes across the full severity range as the first step in generating rear-end crash scenarios. ...
The use of virtual safety assessment as the primary method for evaluating vehicle safety technologies has emphasized the importance of crash scenario generation. One of the most common crash types is the rear-end crash, which involves a lead vehicle and a following vehicle. Most studies have focused on the following vehicle, assuming that the lead vehicle maintains a constant acceleration/deceleration before the crash. However, there is no evidence for this premise in the literature. This study aims to address this knowledge gap by thoroughly analyzing and modeling the lead vehicle’s behavior as a first step in generating rear-end crash scenarios. Accordingly, the study employed a piecewise linear model to parameterize the speed profiles of lead vehicles, utilizing two rear-end pre-crash/near-crash datasets. These datasets were merged and categorized into multiple sub-datasets; for each one, a multivariate distribution was constructed to represent the corresponding parameters. Subsequently, a synthetic dataset was generated using these distribution models and validated by comparison with the original combined dataset. The results highlight diverse lead-vehicle speed patterns, indicating that a more accurate model, such as the proposed piecewise linear model, is required instead of the conventional constant acceleration/deceleration model. Crashes generated with the proposed models accurately match crash data across the full severity range, surpassing existing lead-vehicle kinematics models in both severity range and accuracy. By providing more realistic speed profiles for the lead vehicle, the model developed in the study contributes to creating realistic rear-end crash scenarios and reconstructing real-life crashes.
... However, the THW also determines the situational urgency in a lead vehicle deceleration event. As proved by Wang et al. (Wang et al. 2016), the increment of THW increased situational urgency, thus lead to an reduced BRT to longitudinal conflict. Therefore, the compensated THW in cognitive load may cause an increased BRT which has not been considered in previous studies. ...
... To measure participants' gaze concentration in the process of conflict perception, gaze dispersion was calculated based on raw gaze data belongs to fixation and saccade (Li et al. 2018) in the period of 10 s before the onset of lead vehicle deceleration event. In addition, initial THW (i.e., the THW at the beginning of lead vehicle deceleration event), which reflects both driver's risk compensatory behavior and situational urgency of the conflict (Wang et al. 2016, Young et al. 2007) was extracted. After that, the influence of cognitive load on both gaze dispersion and initial THW was tested by paired-samples t-test. ...
... Most car-following models have used a single reaction time parameter for all car-following conditions (Treiber & Kesting, 2013). However, past studies found that traffic kinematics are correlated with level of urgency, which determines the start of the reaction (Wang et al., 2016;Elhenawy et al., 2017;Wu and Lin, 2019). It is expected that the level of urgency is generally lower in approaching condition (transition from free-driving to approaching a slow lead vehicle at large spacing) than braking condition (transition from closely following to braking). ...
... For instance, Lee (1976) proposed a threshold of s = 8 s for a comfortable deceleration of À3 m/s 2 . Some studies also showed the negative effect of level of urgency (based on time-to-collision) on the start time of brake reaction (Wang et al., 2016;Elhenawy et al., 2017;Wu and Lin, 2019). In contrast, the Wiedemann model does not use a looming variable to predict the emergency braking. ...
... As shown in Figure 9, when the cumulative operating frequency of throttle release reached 50%, the cumulative operating frequency of braking operation only reached about 15%. At this time, the vehicle travelled to the position of about 30 m in front of the taper starting point, which showed that when different drivers faced a low urgency situation, the transition time between throttle release and braking was longer [9]. When the vehicle was at the beginning of the deceleration lane, the cumulative operating frequency of throttle release reached 85%, while the braking cumulative operating frequency was only about 50%. ...
... 190 m downstream of the taper starting point), the cumulative operating frequency of braking only reached 85%. At this time, the cumulative operating frequency of the throttle release reached almost 100%, which also showed that with the driving situation becoming increasingly urgent, drivers in different groups released the throttle and braked more frequently [9]. It is worth noting that when the cumulative frequency of braking reached 75%, the braking operations by drivers in different groups began to become consistent. ...
The deceleration lane isbefore numbers to minus signs. a critical part of the freeway, enabling vehicles to exit the expressway safely and in an orderly fashion. However, drivers are human and thus make subjective decisions while driving; as such, each driver may approach and traverse the deceleration lane differently. This variance, which can cause major traffic disruptions and collisions, can be observed – and even mitigated, as proposed herein – based on a driver's particular characteristics. To study the variances in the longitudinal vehicle positions and microscopic operating characteristics of different drivers on a freeway exit area, a field operational test involving 46 subjects was carried out to collect data on driver characteristics, vehicle motion postures, micro-driving operations, and road geometric elements. The 46 participants were observed based on experience and gender, and the mathematical statistical method was used to analyse driving differences in the deceleration lane. The results show that (1) the vehicle motion state can be divided into four operational stages on the deceleration lane: the pre-deceleration process, the dynamic adjustment process, the first braking process, and the second braking process; (2) drivers generally adopted deceleration behaviour rather than maintain uniform speed when driving in the taper; (3) about 50% of drivers braked after entering the deceleration lane; (4) male drivers and skilled drivers were more inclined to drive at a higher speed on the deceleration lane, and female drivers showed a sharp increase in braking frequency once they travelled 110–150 m downstream of the taper starting point. The results of this study provide data and insights for deceleration lane design, traffic management, and driver training. © 2021 The Authors. IET Intelligent Transport Systems published by John Wiley & Sons Ltd on behalf of The Institution of Engineering and Technology
... Dynamic traffic situations including THW (or following distance), time to collision (TTC), speed, lead vehicle deceleration, and lead vehicle brake light conditions have been proved to affect driver's response to a conflict as well as the crash risk [9,[31][32][33][34]. It has been commonly recognized that the situation urgency increases with the decrease of THW and TTC, and the increase of lead vehicle deceleration and speed [31,34]. ...
... Dynamic traffic situations including THW (or following distance), time to collision (TTC), speed, lead vehicle deceleration, and lead vehicle brake light conditions have been proved to affect driver's response to a conflict as well as the crash risk [9,[31][32][33][34]. It has been commonly recognized that the situation urgency increases with the decrease of THW and TTC, and the increase of lead vehicle deceleration and speed [31,34]. As a consequence, driver brakes faster because of the enhanced visual looming cue [32], and the crash risk is increased due to the increased situation urgency. ...
Abstract Human drivers conduct compensatory behaviour to counteract the increased risk while being distracted. This kind of compensation strategy should be learned for a safer and smart design of adaptive cruise control system (ACC). Hence, a simulator study was conducted, requiring performance visual, cognitive, and combined secondary tasks during a car following scenario. An increased time headway (THW) was found in all of the three distraction conditions, which confirmed that drivers compensated an extra THW to counteract the increased crash risk. Furthermore, crash probability models using binary logistic regression with random intercept were constructed where driver distraction and dynamic traffic situations were embodied as inputs. Results showed that crash risk increased with reduced THW, increased lead vehicle deceleration, and unopened brake light of the lead vehicle. Besides, visual‐related distractions increased crash risk, while pure cognitive distraction lowered crash risk in low THW (lower than 1.8 s) condition and increased crash risk in high THW (larger than 1.8 s) condition. Based on the authors' proposed models, theoretical compensation in THW to fully counteract the increased crash risk by distraction was derived, which could be used for the design of a human‐like ACC with automatic adjustment of THW setting considering driver distraction.
... Car-following behavior has significant impacts on traffic flow efficiency and safety, particularly concerning rear-end crashes Zhang et al., 2019). According to estimates, rear-end crashes account for about 20-30 % of all road traffic crashes (RTCs) and around 10 % of all fatal RTCs (Wang et al., 2016). ...
... The runtime ranged from 0.031 s to 0.036 s, exhibiting poor detecting accuracy, while it reached 0.7 s at its highest precision. Previous studies on the PRT of the adaptive cruise control (ACC) system (Wang et al. 2016;Zhu et al. 2020) have shown a common PRT varying from 0.5 to 1.0 s, slightly shorter than that of humans. In studies focusing on quantifying ODD for AVs (Khoury et al. 2019;Qin et al. 2022;Urmson 2006;Wang et al. 2021;Ye et al. 2021;Ye and Wang 2022), the PRT was commonly set at 0.5 s to calculate the required SSD for AVs at intersections and entrance terminals. ...
With the advancement of automated driving technology, ensuring the operational safety of automated vehicles (AVs) has become a research focus. AVs at freeway exit terminals face high risks due to diverging operations and limited sight distance. Currently, exit terminal design relies on sight distances adequate for human drivers' perception abilities. However, since detecting angle and range are key parameters in the sight triangle and differ between AVs and human drivers, the existing design may not adequately accommodate AVs. To address this gap, this research adopts a strictly mathematical approach to compute operational design domain (ODD) constraints based on the sight triangle at exit terminals. Required stopping sight distance (SSD) or required detecting range, and required detecting angle were quantified as ODD indexes for diverging AVs at various exit ramp and through-lane design speed combinations, based on vehicle kinematics theorems and the sine law. Both flat grades and grades over 3% were considered. Results indicate that (1) the largest required detecting range and angle are 303.0 m and 50.45 degrees, respectively; and (2) some diverging AVs may fail to meet the detecting range requirements, particularly at higher design speeds. Three contributions are provided: (1) the computed ODDs can serve as general references for ensuring sight distances and designing deceleration lanes at exit terminals suitable for AVs; (2) the criteria can improve the geometric design of exit terminals within usual ODD constraints; and (3) it benefits AVs' ODD management by governments and traffic departments.
... It was found that drivers involved in skidding crashes showed significant acceleration and deceleration activities. Similarly, Wang et al. [45] used a high-fidelity driving simulator to investigate the effects of situational urgency on drivers' collision avoidance behaviors and observed that drivers quickly released the accelerator and applied maximum braking force with increasing situational urgency. ...
Truck skidding crashes on horizontal curves pose a significant road safety concern, with improper braking being the primary cause. A data- and model-integrated driven method is proposed to investigate the mechanism and recommend the maximum safe braking deceleration rates without skidding (abbreviated as MSBDRs) for trucks on horizontal curves. Firstly, a comprehensive road–vehicle interaction model was developed, considering dynamic changes in brake force distribution, vertical tire load, and longitudinal and side friction during braking. Secondly, leveraging the “HighD” data set and employing cluster analysis principles, parameter data were extracted using Python and Matlab. Finally, through parameterizing model inputs, the transient dynamic response of trucks was examined, the potential of truck skidding was predicted, and the MSBDRs were recommended. The results indicate the following. (1) There is little concern of truck skidding during car-following braking maneuvers; however, there is a high potential of truck skidding during emergency braking maneuvers. (2) The MSBDR is 4.5 m/s² on a limit-minimum-radius horizontal curve; however, when combined with steep slopes, an overspeed exceeding 20%, and extremely wet road conditions, respectively, the MSBDRs decrease to 4 m/s², 3 m/s², and 2 m/s². These results provide a theoretical foundation for braking strategies in autonomous vehicles.
... Accurately modeling car-following behavior is crucial for microscopic traffic simulation and plays a key role in adaptive cruise control (ACC) systems. Over the years, researchers have shown significant interest in developing car-following models (1)(2)(3)(4)(5) to simulate and understand the dynamics of car-following. ...
Car-following (CF) modeling, a fundamental component in microscopic traffic simulation, has attracted increasing interest of researchers in the past decades. In this study, we propose an adaptable personalized car-following framework -MetaFollower, by leveraging the power of meta-learning. Specifically, we first utilize Model-Agnostic Meta-Learning (MAML) to extract common driving knowledge from various CF events. Afterward, the pre-trained model can be fine-tuned on new drivers with only a few CF trajectories to achieve personalized CF adaptation. We additionally combine Long Short-Term Memory (LSTM) and Intelligent Driver Model (IDM) to reflect temporal heterogeneity with high interpretability. Unlike conventional adaptive cruise control (ACC) systems that rely on predefined settings and constant parameters without considering heterogeneous driving characteristics, MetaFollower can accurately capture and simulate the intricate dynamics of car-following behavior while considering the unique driving styles of individual drivers. We demonstrate the versatility and adaptability of MetaFollower by showcasing its ability to adapt to new drivers with limited training data quickly. To evaluate the performance of MetaFollower, we conduct rigorous experiments comparing it with both data-driven and physics-based models. The results reveal that our proposed framework outperforms baseline models in predicting car-following behavior with higher accuracy and safety. To the best of our knowledge, this is the first car-following model aiming to achieve fast adaptation by considering both driver and temporal heterogeneity based on meta-learning.
... Ref. [17] investigated how safety messages influence driving behavior and proposed that identical messages could yield contrasting outcomes depending on the context. Ref. [18] investigated the crash avoidance behavior of drivers under varying degrees of situational urgency was investigated by means of a high-fidelity driving simulator. Ref. [19] provided evidence that both the onset and regulation of braking were influenced by the time to collision (TTC) when the preceding vehicle initiated braking across various time-headway (THW) conditions. ...
The contribution of this paper is to present an automatic emergency braking (AEB) optimized algorithm based on time to collision (TTC) and a professional driver fitting (PDF) braking pattern. When the TTC value is less than the given threshold, the PDF control algorithm will be started, and vice versa. According to the standard test scenarios for passenger cars and commercial vehicles, the simulation analysis on the AEB systems using four different control algorithms, namely TTC, quadratic curve deceleration, PDF and proposed optimized control algorithm, is conducted, respectively. The results show that the proposed optimization algorithm can both meet the standard requirements and improve the ride comfort. While ensuring collision avoidance with the preceding vehicle, the control algorithm proposed in this study offers better braking comfort compared to the TTC algorithm and the quadratic curve deceleration algorithm. Additionally, it provides a more appropriate stopping distance compared to the PDF algorithm.
... It involves actions taken by a driver when following another vehicle ahead. Proper car-following behavior reduces crashes and improves traffic flow stability [1][2][3][4] . The corresponding car-following model is a mathematical or computational representation of the behavior exhibited by drivers when following other vehicles on the road. ...
Car-following is a control process in which a following vehicle adjusts its acceleration to keep a safe distance from the lead vehicle. Recently, there has been a booming of data-driven models that enable more accurate modeling of car-following through real-world driving datasets. Although there are several public datasets available, their formats are not always consistent, making it challenging to determine the state-of-the-art models and how well a new model performs compared to existing ones. To address this gap and promote the development of microscopic traffic flow modeling, we establish the first public benchmark dataset for car-following behavior modeling. This benchmark consists of more than 80 K car-following events extracted from five public driving datasets under the same criteria. To give an overview of current progress in car-following modeling, we implemented and tested representative baseline models within the benchmark. The established benchmark provides researchers with consistent data formats and metrics for cross-comparing different car-following models, coming with open datasets and codes.
... It involves actions taken by a driver when following another vehicle ahead. Proper car-following behavior can lower crash risks and improve traffic flow stability [1]- [4]. The corresponding car-following model is a mathematical or computational representation of the behavior exhibited by drivers when following other vehicles on the road. ...
Car-following is a control process in which a following vehicle (FV) adjusts its acceleration to keep a safe distance from the lead vehicle (LV). Recently, there has been a booming of data-driven models that enable more accurate modeling of car-following through real-world driving datasets. Although there are several public datasets available, their formats are not always consistent, making it challenging to determine the state-of-the-art models and how well a new model performs compared to existing ones. In contrast, research fields such as image recognition and object detection have benchmark datasets like ImageNet, Microsoft COCO, and KITTI. To address this gap and promote the development of microscopic traffic flow modeling, we establish a public benchmark dataset for car-following behavior modeling. The benchmark consists of more than 80K car-following events extracted from five public driving datasets using the same criteria. These events cover diverse situations including different road types, various weather conditions, and mixed traffic flows with autonomous vehicles. Moreover, to give an overview of current progress in car-following modeling, we implemented and tested representative baseline models with the benchmark. Results show that the deep deterministic policy gradient (DDPG) based model performs competitively with a lower MSE for spacing compared to traditional intelligent driver model (IDM) and Gazis-Herman-Rothery (GHR) models, and a smaller collision rate compared to fully connected neural network (NN) and long short-term memory (LSTM) models in most datasets. The established benchmark will provide researchers with consistent data formats and metrics for cross-comparing different car-following models, promoting the development of more accurate models. We open-source our dataset and implementation code in https://github.com/HKUST-DRIVE-AI-LAB/FollowNet.
... Due to the low sample size in each driver category, the interaction effect of ADB and conflict approaching time gaps on crash risk is not significant. However, to analyze the main effects of ADB and conflict approaching time gaps (aggressive (23 × 5), moderately aggressive (16 × 5), and non-aggressive drivers (16 × 5)) the sample size is sufficient to emphasize the valid findings based on previous studies (Li et al., 2019b;Wang et al., 2016;Yang et al., 2021). Nevertheless, future studies can improve the sample size by considering a better representation of all driver categories. ...
Different levels of aggressive drivers are identified based on multi-dimensional vehicle kinematics.
A validation of the driver detection method is performed using self-reported trait and instrumental aggression to ensure the accuracy of the driver detection method.
The effect of aggressive driver behavior on crash risk is quantitatively analyzed based on time to collision (TTC) during an emergent pre-crash scenario.
Further, collision evasive behavior is analyzed based on speed reduction time (SRT) survival probabilities.
... However, an effective FCW system should be designed on the base of a good understanding of the collision avoidance behaviour. There is considerable literature supporting the importance of a quick response to the lead vehicle's brake and major factors that may affect drivers' brake response time, including drivers' age, gender, profession, weather condition, and so on [4][5][6][7][8][9][10][11]. As for drivers' crash avoidance behaviour after brake onset, both driving simulator studies and naturalistic studies demonstrated that drivers often reached a maximum deceleration rate [12,13] and there was a stepwise ramp-up towards the maximum deceleration [6,12,13]. ...
Considering the large proportion of rear-end collisions occurring in our daily life and the severity it may lead to, the objective of this study was to investigate the effect of situation kinematics on drivers’ rear-end collision avoidance behaviour after brake onset. A wide range of lead vehicle deceleration scenarios were designed based on driving simulator experiments to collect drivers’ deceleration behaviour data. Different from measures (e.g., speed, the lead vehicle’s deceleration et al.) often adopted in previous studies, a visual looming-based measure at different time points was calculated combined with analysis of speed and distance to quantify situation kinematics in this study. The Spearman’s nonparametric rank correlation test was firstly conducted to examine the correlation between visual looming-based metrics and related deceleration behaviour. The mixed model was performed on drivers’ brake jerk and maximum deceleration rate, while the logistic model was then performed to predict the probability of the occurrence of rear-end collisions. Spearman’s nonparametric test showed that both deceleration ramp-up and drivers’ maximum deceleration rate increase significantly as the looming traces increase faster. Results of the logistic model indicated that the probability of occurrence of a potential collision might be higher if the situation at the brake onset is quite urgent and braking is moderate. It was demonstrated that both drivers’ deceleration ramp-up and maximum deceleration rate could be highly kinematic-dependent, and visual looming, driving speed, and distance can be useful information for drivers to take relative deceleration actions.
... Consequently, there is a need to understand driving skills, and how they are affected by stresses to respond quickly that occur in normal driving. Previous research has shown that drivers' responses to a decelerating car were more acute (e.g., braking hard to a full stop) as the distance between the cars decreased (Wang et al., 2016). While these acute responses to unexpected events are crucial for avoiding collisions or minimizing damage (Harb et al., 2009), the type of response chosen by the driver might not always be the optimal one (Markkula et al., 2012). ...
In everyday driving on the road, people are often required to make fast decisions that could compromise the accuracy of choices. We present a diffusion model analysis of the adjustments drivers make to the decision process under speed-stress. Participants operated a PC-based driving simulator while performing one of two decision-making tasks that required a driving action as a response to the stimulus. In a one-choice driving task, participants were asked to drive around a lead car when its brake lights were turned on. A two-choice driving task used a brightness-discrimination task in which participants were asked to drive to the left and back behind a lead car if there were more black than white pixels in a display and to the right and back if there were more white than black pixels. Speed-stress was operationalized by instructing drivers to respond as quickly as possible and by manipulating the distance drivers were required to maintain behind the lead car. Results showed the expected speed-accuracy tradeoff; however, the cost on accuracy in the two-choice task was relatively small. The model-based analysis showed that this was achieved by lowering the decision criteria and speeding up nondecision processes without disrupting components that produce evidence for the decision process. In fact, in the one-choice task, evidence accumulation rate in the speed-stress condition was found to be higher than in the accuracy-stress condition. We concluded that drivers were able to comply with speed-stress demands with relatively safe adjustments that imposed minimal costs on the accuracy of choices.
... However, Meng et al. [21] found that trucks have a much higher probability of being involved in a rear-end crash than a car at the work zone. For the rear-end collision situation, as situational urgency increased, drivers released the accelerator and braked to maximum more quickly [22]. ...
A large amount of traffic crash investigations have shown that rear-end collisions are the main type collisions on the freeway. The purpose of this study is to investigate the rear-end collision risk on the freeway. Firstly, a new framework was proposed to develop the rear-end collision probability (RCP) model between two vehicles based on Generalized Pareto Distribution (GPD). Secondly, the freeway rear-end collision risk (F-RCR) was defined as the sum of the rear-end collision probability of each vehicle and divided into three levels which was high, median, and low rear-end collision risk. Then, different machine learning algorithms were used to model F-RCR under the condition of an unbalanced dataset. The result of the RCP model showed continuous change and can identify the dangerous condition quickly compared to the traditional models even when the speed of the leading vehicle is faster than the following vehicle. When the vehicle distribution was unbalanced on road and the speed difference between adjacent lanes and the traffic volume was large, F-RCR will increase. Multi-Layer Perceptron (MLP) was found to be more suitable for modeling F-RCR. The framework provided in this research was transferrable and can be used in the freeway proactive traffic safety management system.
... According to the National Highway Traffic Safety Administration, rear-end collisions in the United States constituted 32.5% of all crashes in 2019 [2]. In Shanghai, China, approximately 20% of all road crashes were rear-end collisions: 49% are elevated expressway collisions, and 67% are tunnel collisions [3]. erefore, rearend collisions must be prevented to improve road traffic safety. ...
Vehicle rear-end collisions are primarily caused by tight car following in a continuous traffic flow, as well as a driver’s incorrect perception of the traffic environment ahead and delayed response. To facilitate an investigation pertaining to rear-end collision mechanisms and accurately measure the risk, the concept of a vehicle group is introduced herein. A risk measurement model for a vehicle group (RMVG) based on temporal and spatial similarities is proposed. First, vehicles are categorized based on their temporal and spatial similarities. Risk measurement metrics are defined based on the traffic composition, movement state, and conflict extent. Subsequently, vehicle group risk identification and risk measurement models based on an isolation forest are established. the rear-end collision risk of the vehicle groups is analyzed both qualitatively and quantitatively. Finally, the RMVG is tested using the vehicle trajectory data set of Longpan South Road, Nanjing City, Jiangsu Province, China, and the results are compared with those of a support vector machine and local outlier factor. The results show that the accuracy of the RMVG is higher than those of other models: its accuracy rate and specificity are 95.68% and 88.89%, respectively, whereas its false alarm rate is only 3.47%.
... The value range of the velocity and the distance gap were selected based on the parameters' true distributions of Shanghai expressway system. Different deceleration values of HDV1 were set according to the literature [41]. ...
Highly automated vehicles (HAVs) have been introduced to the transportation system for the purpose of providing safer mobility. Considering the expected long co-existence period of HAVs and human-driven vehicles (HDVs), the safety operation of HAVs interacting with HDVs needs to be verified. To achieve this, HAVs’ Operational Design Domain (ODD) needs to be identified under the scenario-based testing framework. In this study, a novel testing framework aiming at identifying the Safety performance boundary (SPB) is proposed, which assures the coverage of safety-critical scenarios and compatible with the black-box feature of HAV control algorithm. A surrogate model was utilized to approximate the safety performance of HAV, and a gradient descent searching algorithm was employed to accelerate the search for SPB. For empirical analyses, a three-vehicle following scenario was adopted and the Intelligent Driver Model (IDM) was tested as a case study. The results show that only 4% of the total scenarios are required to establish a reliable surrogate model. And the gradient descent algorithm was able to establish the SPB by identifying 97.42% of collision scenarios and only false alarming 0.29% of non-collision scenarios. Furthermore, the concept of safety tolerance was proposed to measure the possibilities of boundary scenarios dropping in safety performance. The applications of helping to construct ODD and compare different control algorithms were discussed. It shows that the IDM performs better than the Wiedemann 99 (W99) model with larger ODD.
... Rear-end collision is one of the top two collision patterns on freeways in the world. According to the NHTSA (National Highway Traffic Safety Administration) traffic accident data in 2014-2018, rear-end collisions accounted for the highest proportion, reaching 40% (NHTSA, 2018) and about 10% of all fatal crashes (Wang et al., 2016a) in the United States. In China, the 2018 traffic accident data revealed that rear-end collisions accounted for the second-highest proportion of 7.95% (Institute of Highway Science, 2019). ...
Rear-end collision and side collision are two types of accidents with the highest accident rate in the world. Numerous studies have focused on rear-end accident research, but only a few constructive countermeasures are put forward. Driving risk evolution at the driver operational level before an accident is critical to collision avoidance. This paper puts forward a driver operational level identification of driving risk and graded alarm under near-crash conditions. Firstly, driving simulation is utilized to acquire the operation data of SV (subjective vehicle) under the condition of emergent deceleration of LV (leading vehicle). The kinematic model is built to characterize the law of the risk discrimination indices of SV including THW (time headway), SHW (space headway) and TTCi (the reciprocal of time to collision). The predicted results are consistent with the naturalistic driving data. Secondly, the three-dimensional distribution 'speed-spacing-TTCi' is applied to classify the risky driving state of SV. The precarious distribution is concentrated at the area where relative velocity increased to 23-40 km/h and spacing decreased to 18-30 m. Finally, based on the reaction time and braking distance reduction, the optimal external intervention is determined to be the acousto-optic way by driving simulation. For moderate drivers, a three-level alarm of 2.94 s, 1.94 s and 1.1 s is calibrated considering different driving styles and cumulative frequency curve of reaction time.
... Traffic crash is a serious problem in the current transportation system, leading to heavy casualties and economic losses [1][2]. Among all traffic crashes, rear-end crashes account for approximately 32% in the United States and 20% in Shanghai, China [3]. Crash statistics show that most rear-end crashes are caused by improper operation and delayed reaction of drivers [4][5][6]. ...
Cooperative adaptive cruise control is recognised as an effective means to reduce the rear-end collision caused by the delayed reaction and improper driving behaviours of manually driven vehicles. However, vehicle-to-vehicle communication, as the key to collect information for the cooperative adaptive cruise control system, is likely to be interrupted by malicious attacks and equipment failures. When communication interruption occurs, sensors of the adaptive cruise control system will replace vehicle-to-vehicle communication for information collection, that is, cooperative adaptive cruise control vehicles degrade to adaptive cruise control vehicles automatically, causing an increased rear-end collision risk. This study aims to investigate the safety impact of cooperative adaptive cruise control vehicles’ degradation under spatial continuous communication interruption. A mixed platoon, consisting of cooperative adaptive cruise control vehicles, adaptive cruise control vehicles, manually driven vehicles, and improved manually driven vehicles with information-sending devices, is simulated with the realistic cooperative adaptive cruise control and adaptive cruise control model, and the intelligent driver model. Different platoon forms are discussed by adjusting the quantity and position of manually driven vehicles in the mixed platoon. The results show that manually driven vehicles can effectively block the backward extension of speed fluctuation and reduce the rear-end collision risk under the spatial continuous communication interruption. Nevertheless, equipping manually driven vehicles with information-sending devices weakens the ability to defend against communication interruption. © 2021 The Authors. IET Intelligent Transport Systems published by John Wiley & Sons Ltd on behalf of The Institution of Engineering and Technology
... The runtime lasted for 0.7 s for an accurate detector aiming at maximum detecting accuracy, and varied from 0.031 s to 0.036 s for fast detectors with low average detection precision. As for the PRT of the adaptive cruise control (ACC) system, especially in commercial vehicles, was found to be slightly shorter than that of human drivers, with values from 0.5 to 1.0 s demonstrated in previous studies (Wang et al., 2016;Zhu et al., 2020). Additionally, the PRT parameter was frequently calibrated to be 0.5 s in studies on the responsibility-sensitive safety (RSS) model (Liu et al., 2021;Xu et al., 2021). ...
In recent years, the development and testing of autonomous driving technology have become widespread around the world. However, due to differences in perception abilities between autonomous vehicles and human drivers, the current geometric design controls for highway alignments, designed for the human driver, may not be applicable to the autonomous vehicle (AV). Few studies, however, have systematically investigated the design controls for autonomous vehicles, though we face full driving automation in the next few decades. Because the range of modern AV sensors reaches 250 m, with expected further improvements in the near future, there is a need to determine how the sensors’ perception field and perception-reaction time may affect the current road design standards developed for human drivers. This study therefore tested the feasibility of the current design controls for fully-autonomous vehicles by separately computing controls for vertical alignments and combined horizontal and vertical alignments, considering the AV’s perception abilities of perception-reaction time (PRT), sensor height, and upward angle from the horizontal. The required stopping sight distance (SSD) and minimum length of sag and crest vertical curves were derived and compared with those for human drivers. Computations for combined alignments were based on Green Book coordination guidelines: as the minimum length of horizontal curve can be used for alignments adhering to guidelines, preview sight distance (PVSD) was computed for alignments that do not. Results showed that 1) AV-based design controls on vertical curves were more tolerant than those based on human drivers; and 2) the dominating criterion of sag vertical curve design control was comfort for autonomous vehicles, versus required SSD for human drivers.
Extensive experimental analyses concerned with Adaptive Cruise Control (ACC) have clearly shown that such systems have failed to deliver the promise of safe and traffic-flow effective car-following. On the contrary, large reaction times and poor string stability performances characterize commercial ACCs. While a huge research line is investigating the introduction of communication among vehicles to overcome the mentioned limitation, market adoption of connectivity-enhanced vehicles is struggling. In this context, an alternative approach based on multiple vehicle anticipation using RADAR only has emerged. Multianticipation is definitely not a new concept within the transportation community. However, until now, it was mainly associated with human driving. In the present manuscript, we demonstrate instead how, at least, one vehicle manufacturer has implemented multianticipation on a commercial vehicle. Following an in-house carried out testing campaign, we give an experimental characterization of the functioning of such a system including the potential impact on the flow and safety using a state-of-the-art fuzzy-logic safety performance model. The first results demonstrate that the vehicle under test reacted to one additional vehicle in front of the leader vehicle. Moreover, the actual realization appears to mainly target safety applications whereas there is only a marginal benefit on the string stability characteristics of the system. While we recorded a marginal string stability improvement (about 10 %), the minimum TTC was twice as large when multianticipation occurred with respect to the cases when that was not activated. Relevant Fuzzy Surrogate Safety Metrics further supported the safety argument.
In the realm of driving technologies, fully autonomous vehicles have not been widely adopted yet, making advanced driver assistance systems (ADAS) crucial for enhancing driving experiences. Adaptive Cruise Control (ACC) emerges as a pivotal component of ADAS. However, current ACC systems often employ fixed settings, failing to intuitively capture drivers' social preferences and leading to potential function disengagement. To overcome these limitations, we propose the Editable Behavior Generation (EBG) model, a data-driven car-following model that allows for adjusting driving discourtesy levels. The framework integrates diverse courtesy calculation methods into long short-term memory (LSTM) and Transformer architectures, offering a comprehensive approach to capture nuanced driving dynamics. By integrating various discourtesy values during the training process, our model generates realistic agent trajectories with different levels of courtesy in car-following behavior. Experimental results on the HighD and Waymo datasets showcase a reduction in Mean Squared Error (MSE) of spacing and MSE of speed compared to baselines, establishing style controllability. To the best of our knowledge, this work represents the first data-driven car-following model capable of dynamically adjusting discourtesy levels. Our model provides valuable insights for the development of ACC systems that take into account drivers' social preferences.
The study investigates the impact of driving aggression on lane change behavior, especially during critical merging gaps. We have observed notable differences in maneuvering dynamics among drivers with varying levels of driving aggression, including decision-making patterns, compensatory steering behavior, and lane change execution duration. Aggressive drivers tend to complete lane changes significantly faster, often resorting to extreme steering behaviors such as increased steering reversal rates and steering swerves. Additionally, we have analyzed the consequences of reckless maneuvering on the likelihood of crashes. Despite employing extreme steering efforts to compensate for their reckless driving and avoid collisions, these drivers experienced significantly higher crash risk during lane changes.
Find the article here: htps://authors.elsevier.com/c/1j3vE4tTwCyYVf
The severity of vehicle-pedestrian crashes has prompted authorities worldwide to concentrate on improving pedestrian safety. The situation has only become more urgent with the approach of automated driving scenarios. The Responsibility-Sensitive Safety (RSS) model, introduced by Mobileye®, is a rigorous mathematical model developed to facilitate the safe operation of automated vehicles. The RSS model has been calibrated for several vehicle conflict scenarios; however, it has not yet been tested for pedestrian safety. Therefore, this study calibrates and evaluates the RSS model for pedestrian safety using data from the Shanghai Naturalistic Driving Study. Nearly 400 vehicle-pedestrian conflicts were extracted from 8,000 trips by the threshold and manual check method, and then divided into 16 basic scenarios in three categories. Because crossing conflicts were the most serious and frequent, they were reproduced in MATLAB's Simulink with each vehicle replaced with a virtual automated vehicle loaded with the RSS controller module. With the objectives of maximizing safety and minimizing conservativeness, the non-dominated sorting genetic algorithm II was applied to calibrate the RSS model for vehicle-pedestrian conflicts. The safety performance of the RSS model was then compared with that of the commonly used active safety function, autonomous emergency braking (AEB), and with human driving. Findings verified that the RSS model was safer in vehicle-pedestrian conflicts than both the AEB model and human driving. Its performance also yielded the best test results in producing smooth and stable driving. This study provides a reliable reference for the safe control of automated vehicles with respect to pedestrians.
A necessary component of any crash avoidance or stopping distance calculation is the perception-response time. Courts have excluded the testimony of experts who used generalized average response times. Throughout history, reaction time research demonstrated that response times changed based upon probability of the event and other factors. This research will examine how drivers' response times change based upon the comparative probability of the event. The goal of the research is to determine the average and distribution of response times for several common path intrusion events. After an analysis of the related literature, four primary factors were found to be associated with drivers' response times when responding to path intrusion events. 1. Comparative probability of the event was correlated with the response time. 2. Measured response times were longer when the onset (i.e., start of the measurement) was earlier in the event, and shorter when the onset was later in the event 3. The difference between brake response time (i.e., when braking occurred) and perception-response time (i.e., when hard braking or sharp steering began) was approximately 0.3 seconds. 4. For events other than very common and very rare, the coefficient of variation, which is the standard deviation divided by the mean was within 0.4 second. There was a wider distribution of response times for shorter BRT events such as cutoffs , sudden appearance from behind obstacles, and longer BRT events such as head-ons (SD/Mean = 0.54). This paper will identify common mistakes made by crash investigators and then describe a classical scientific approach that meets the requirements of any court and eliminates bias from the analysis. The classical approach entails comparing the pre-impact response of a driver against the range of typical driver and rider responses in research when faced with a similar response scenario.
In recent years, many studies have used poor cognitive functions to explain risk safety differences among drivers. Working memory is a cognitive function with information storage and attentional control that plays a crucial role in driver information processing. Furthermore, it is inextricably linked to parameters such as driving performance, driving eye movements and driving neurophysiology, which have a significant impact on drivers' risky behavior and crash risk. In particular, crash risk is a serious risk to social safety and economic development. For this reason, it is necessary to understand how risk-related working memory affects driving so that pre-driving safety pre-training programs and in-vehicle safety assistance systems for driving can be developed accordingly, contributing to the development of semi-autonomous vehicles and even autonomous vehicles. In this paper, a systematic search of the literature over the past 23 years resulted in 78 articles that met the eligibility criteria and quality assessment. The results show that higher working memory capacity, as measured neuropsychologically, is associated with more consistent and safer driving-related parameters for drivers (e.g., lane keeping) and may be related to pupil dilation during risk perception while driving, which is associated with driving outcomes (tickets, pull-overs, penalty points and fines,and driving accidents) is closely related to the perceived usefulness of the human-machine interface, reaction time, standard deviation of steering wheel corners, etc. when the autonomous driving takes over. In addition, higher working memory load interference was associated with more inconsistent and unsafe driving-related parameters (including but not limited to eye movements, electrophysiology, etc.), with higher working memory load being associated with easier driver concentration on the road, faster heart rate, lower heart rate variability, and lower oxyhemoglobin (OxyHb) and deoxyhemoglobin (DeoxyHb). Only a limited number of studies have simultaneously investigated the relationship between working memory capacity, working memory load and driving, showing an interaction between working memory capacity and working memory load on lane change initiation and lane change correctness, with working memory capacity acting as a covariate that mediated the effect of working memory load on braking reaction time. In addition, working memory-related cognitive training had a transfer effect on improving driving ability. Overall, working memory capacity determines the upper limit of the number of working memory attention resources, while working memory load occupies part of the working memory attention resources, thus influencing information perception, decision judgment, operational response, and collision avoidance in driving. Future effective interventions for safe driving can be combined with capacity training and load alerting. These findings contribute to our understanding of the role of working memory in driving and provide new insights into the design of driver safety training programs and automated driving personalized in-vehicle safety systems and roadside devices such as signage.
In vehicles with driving assistance systems, the responsibility of driving still lies with the driver. Therefore, active safety systems follow a design principle of avoiding intervention as long as possible. Decision making using this principle requires collision avoidance with surrounding objects during the evasive maneuvering of an ego vehicle. However, general decision-making methods for collision avoidance focus on preventing collisions with objects in front of the subject vehicle; the risk of collision with the following vehicle caused by emergency braking is not considered. This paper presents decision-making methods for such collision avoidance systems, wherein emergency braking and steering are considered as collision avoidance maneuvers. The risk of collision is predicted based on the braking model of a normal driver and several emergency braking strategies are designed to avoid collisions. A within-lane steering avoidance method to improve avoidance performance in small overlap collision scenarios and a general avoidance method of steering to change lanes are designed. Collisions with surrounding objects can be avoided using the designed evasive maneuvers; further, maneuvers that can be implemented at the last moment are determined. The results of computer simulations indicate an improved collision avoidance performance using the proposed methods.
In autonomous vehicles (AVs), especially in fully AVs, “drivers” perceive vehicle operation from a passenger’s perspective. This study focuses on the perspective change of drivers especially in fully AVs. To investigate the effect of change in perspective on drivers’ assessment of AVs, this study conducted a driving simulator experiment and a survey using different samples. The driving simulator experiment was a within-subjects design including 40 participants. The experimental results indicated that although AVs drive exactly the same as the drivers, 85% of these drivers had different assessments for AVs compared to driving themselves (self-AV bias). Among these biased drivers, 80% changed the direction of their assessments (e.g., satisfied with their own driving behaviors but dissatisfied with the same behaviors from the AVs). Additionally, self-AV bias increased with the level of self-evaluation of their own driving skills. The higher the level of self-evaluation, the higher the optimism bias (drivers think they drive better than AVs) of drivers. Thereafter, to show whether and how self-AV bias affects drivers’ intuitive acceptance of AVs, a survey was conducted, which included 381 valid questionnaires. The survey results showed that self-AV bias was a critical factor promoting perceived usefulness of AVs; particularly, highly self-AV-biased drivers responded more to the perceived usefulness of AVs. Moreover, drivers who could accept more risky driving behaviors of AVs compared to themselves also perceived AVs to be more useful. The findings from this study provide insights for understanding drivers’ assessment and acceptance of AVs’ driving behaviors.
Driving aggression is a major concern during car-following situations as it is closely associated with collision risk and crash severity. Despite the tendency of reckless driving actions and increased crash risk with the vehicle ahead, aggressive driver behavior remains unexplored. The present study aimed to investigate the effects of aggressive driver behavior (three driver categories: aggressive, moderately aggressive, and non-aggressive drivers) and drivers’ tendency to aggressive stimuli (due to lead vehicle (vehicle moving ahead) reckless driving) on car-following behavior. A sample of fifty-eight Indian drivers participated in this study. All the experiments were conducted using a driving simulator. The simulator design includes a lead vehicle dynamic maneuver with rapid accelerations and decelerations (at 1 m/s², 1.5 m/s², and 2 m/s²). Three longitudinal performance measures including speed variability (SPV) from LV, speed recovery time (SRCT), and time spent tailgating, were analyzed during LV acceleration stage using generalized linear models (GLM). Further, to assess collision avoidance behavior and crash risk, deceleration adjusting time (DAT) was analyzed during the LV sudden deceleration stage using Weibull Accelerated Failure Time (AFT) analysis. The findings indicate that due to their tendency to risk-raking behavior, aggressive drivers decreased their SPV and SRCT by 25 % and 18 %, respectively. The time spent tailgating increased by 107 % for aggressive drivers. The survival probabilities were decreased by 23.77 % for aggressive drivers at 3 sec DAT. The findings of the study can be used in car-following models to enhance the model performance, improve the reliability of simulation tools, and design interventions to improve safety (in automated driving).
The Tampa Hillsborough Expressway Authority Connected Vehicle Pilot Deployment (THEA CV Pilot) implemented several vehicle-to-vehicle (V2V) and vehicle to infrastructure (V2I) applications on more than 1,000 private vehicles. This paper focuses on the Forward Collision Warning (FCW) application to study factors that are associated with drivers’ reactions to FCWs and to investigate if the observed driving styles derived from the data support the participants’ stated driving styles obtained from their survey responses. A panel of participants, driving in real-world traffic conditions for over two years with retrofitted CV technology and integrated FCW application, is used. The panel consists of a treatment (Human Machine Interface (HMI) enabled) and a control (HMI disabled) group. Random parameters logit and correlated grouped random parameter logit models are estimated to reveal possible associations between stated and observed driving behavior, HMI exposure, socio-demographic factors, and the response variable (drivers’ reaction to FCW). The study found an association between one measure of driving volatility, so that with increased driving volatility (proxy for driving aggressiveness), the probability of reaction to FCW declines. The study also found that the probability of reaction for drivers who received a warning (audiovisual) via HMI increased by 9.93 % compared to those who did not receive a warning.
In public road tests of autonomous vehicles in California, rear-end crashes have been the most common type of crash. Collision avoidance systems, such as autonomous emergency braking (AEB), have provided an effective way for autonomous vehicles to avoid collisions with the lead vehicle, but to avert false alarms, AEB tends to apply late and hard brake only if a collision becomes unavoidable. Automatic preventive braking (APB) is a new collision avoidance method used in Mobileye’s Responsibility-Sensitive Safety (RSS) model that aims to reduce crashes with a milder brake and decreased impact on traffic flow, but APB’s safety performance is inferior to that of AEB. This study therefore proposes three safety improvement strategies for APB, the addition of response time, safety buffer, and minimum following distance; and combines them in different ways into four improved APB systems, IP1-IP4. Simulating car-following safety–critical events (SCEs) extracted from the Shanghai Naturalistic Driving Study in MATLAB’s Simulink, the safety performance, conservativeness, and driving comfort of the four systems were evaluated and compared with the original APB system, two AEB systems, and human drivers. The results show that 1) IP4, the system that integrated all three strategies, outperformed the baseline APB and IP1-IP3 and prevented all SCEs from becoming crashes; 2) IP4 was slightly more conservative than AEB, but less conservative than RSS; 3) APB’s jerk-bounded braking profile improved driving comfort; and 4) higher deceleration was found in the two AEB systems (both 8.1 m/s²) than in IP4 (6.7 m/s²), but they failed to prevent all crashes. Our proposed APB system, IP4, can provide safe, efficient, and comfortable braking for AVs in car-following SCEs, and has the potential to be practically applied in vehicle collision avoidance systems.
As the automobile market gradually develops towards intelligence, networking, and information-orientated, intelligent identification based on connected vehicle data becomes a key technology. Specifically, real-time crash identification using vehicle operation data can enable automotive companies to obtain timely information on the safety of user vehicle usage so that timely customer service and roadside rescue can be provided. In this paper, an accurate vehicle crash identification algorithm is developed based on machine learning techniques using electric vehicles’ operation data provided by SAIC-GM-Wuling. The point of battery disconnection is identified as a potential crash event. Data before and after the battery disconnection is retrieved for feature extraction. Two different feature extraction methods are used: one directly extracts the descriptive statistical features of various variables, and the other directly unfolds the multivariate time series data. The AdaBoost algorithm is used to classify whether a potential crash event is a real crash using the constructed features. Models trained with the two different features are fused for the final outputs. The results show that the final model is simple, effective, and has a fast inference speed. The model has an F1 score of 0.98 on testing data for crash classification, and the identified crash times are all within 10 s around the true crash times. All data and code are available at https://github.com/MeixinZhu/vehicle-crash-identification.
The aggregate properties play a major role in the structural and functional performance of various pavement layers. The present chapter provides an introduction to aggregate sources, types, quality, and different size distribution requirements for pavement layers. Further, the chapter is categorized into two broad sections. The first section offers an extensive discussion on past and presently available conventional aggregate characterization practices, limiting values, and their significance. For example, the properties associated with the aggregate source (i.e., abrasion, impact, soundness, and absorption) and consensus or shape (i.e., FAA: fine aggregate angularity; coarse aggregate angularity, flatness and elongation, and sand equivalent) are discussed. The second part of the chapter talks about the application of various image-based techniques for aggregate shape characterization and its futuristic importance for the production of quality aggregates. The Digital Image Technique (DIT)-based methods like University of Illinois Aggregate Image Analyzer (UIAIA), Aggregate Image Measurement System (AIMS), Laser-based Aggregate Analysis System (LAAS), Videographer-40 (VDG-40), and CT-Scan are discussed. Moreover, a brief overview of various algorithms associated with digital image techniques is presented. The part also includes a comparative discussion on conventional and image-based shape characterization approaches. Additionally, the futuristic application of DIT in different aspects like aggregate production, quality control, and pavement forensic investigations are outlined. Overall, it is expected that the chapter will develop a broad understanding among researchers, practitioners, aggregate producers, and pavement stakeholders on the qualitative characterization of aggregates.KeywordsAggregatePavement layersSource propertiesShape propertiesDigital image techniqueCrushing mechanisms
Metropolitan cities of India are recently promoting transit-oriented development (TOD), as a sustainable transport approach for improving urban mobility in the nation. However, TOD policies in Indian cities require a thorough diagnosis of planning parameters based on existing urban structure elements. Existing policies need to enhance and introduce priority planning strategies that adhere to long-term visions of Indian cities. The present chapter pays attention to macro-/micro planning parameters, namely size of influence zone, typology, TOD measurement tool, floor-area-ratio (FAR), building use compositions, and priority strategies that have a substantial effect on planning processes. This chapter specifies quantitative levels of urban structure to assist as a real-world guidebook. It provides a measurement tool to understand the nature of TOD-ness and to establish a standard to plan and implement TOD at neighborhood levels. Such a TOD measurement tool can help planners, policymakers, and government authorities, while investing funds on infrastructure with prior knowledge on existing levels of TOD at neighborhood level. Because without measuring the current situation of TODs, faulty decisions in investments will continue to be repeated.KeywordsTransit-oriented development planningDefinitionTOD policy in IndiaTypologyTOD scoreDelhi Metro
A significant number of fatalities along horizontal curves of rural highways necessitated the safety evaluation in the curve design. Previous studies explored safety evaluation based on several approaches such as operating speed, driver workload, alignment index, and vehicle stability. However, researchers predominantly focused on vehicle speed because it has a close resemblance to design speed and straightforward implementation in the field. The safety evaluation criteria for horizontal curves are majorly speed-based studies. Hence, this chapter represents several operating speed-based safety evaluation approaches available in the literature. It provides insight on the requirement of holistic design guidelines and appropriate safety performance measures to develop safer geometric elements for low-middle income countries like India.
The research conducted on overtaking maneuver for evaluating drivers' safety showed adverse effects of urgency on driving performance and decision making. Therefore, a driving simulator study was designed to examine driving performance of the drivers and its implication on overtaking and crash probabilities under increasing time pressure conditions. Eighty-eight participants data were analyzed in the current study. Three different time pressure conditions: No Time Pressure (NTP), Low Time Pressure (LTP), and High Time Pressure (HTP) were considered for analyzing driving performance of the drivers while executing overtaking maneuvers. The driving performance was assessed using minimum time-to-line crossing and coefficient of variation in speed to dissect the safety margin adopted by the drivers while overtaking the lead vehicle. Further, minimum time-to-line crossing and coefficient of variation in speed were considered as explanatory variables to investigate their influence on overtaking and crash probabilities. Para-metric survival analysis and Generalized Linear Mixed Models (GLMM) were used to assess the driving performance, overtaking and crash probabilities. The parametric survival analysis showed that minimum time-to-line crossing reduced by 36.7% and 63.8% in LTP and HTP driving conditions , respectively. The GLMM results revealed that coefficient of variation in speed increased by 3.437% in HTP (no significant effect in LTP) as compared to NTP driving conditions. Further, the GLMM results showed that overtaking and crash probabilities decreased with increment in minimum time-to-line crossing and coefficient of variation in speed values. Additionally, it was observed that male drivers took risky decisions than female drivers. Nevertheless, the comparative analysis revealed that male drivers were less prone to crashes than female drivers. Overall, it can be inferred that the drivers take risky decisions with increment in time pressure to complete the driving task, even at the expense of their own safety which exposed them to high likelihood of crashes.
Situation assessment is crucial for intelligent vehicles, enabling detection of potential risks to dynamic and complex traffic environments. In this paper, we propose a unified framework that tackles the coupling relationships between traffic participants and quantifies the possible range of vehicle trajectory generation and the expected losses caused by risk source attributes in the driving process. We first apply the state space trajectory planning scheme based on a sampling algorithm to generate the path candidates; each feasible path is designed through a parametric cubic spline. Then, to evaluate the risk range in the driving process, we quantify the interaction of traffic participants, and employ the principle of least action to calculate the cost of each feasible path when achieving the destination. The probability distribution map, namely the possible range of driving trajectories, can be obtained based on the path cost. Furthermore, the vehicle-to-vehicle interaction is calculated based on the equivalent force, which estimates the expected accident losses. Finally, the vehicle trajectory prediction and the expected loss are combined to output the probabilistic situation assessment of intelligent vehicles. The algorithm is implemented in different scenarios and applied to the trajectory planning process. Results demonstrate that, compared with the classical situation assessment metric, the developed method can determine and accurately identify the influence range of driving risk in real-time, predict a dangerous situation earlier, and ensure the vehicle avoids obstacles in advance.
An effective forward collision warning (FCW) system must be compatible with drivers' risk perceptions and behavioral responses. The Collision Avoidance Metrics Partnership (CAMP) developed a kinematic-based FCW algorithm to determine the minimum distance needed to stop safely under various levels of rear-end crash risk. The algorithm generates a linear function for predicting drivers' expected response decelerations (ERDs) by considering motions of the involved vehicles. This linear function works well when the risks perceived by drivers are low; however, at elevated risks when the lead vehicle (LV) decelerates at an unexpectedly high rate, or at high relative speeds, the warnings are triggered too late for the subject vehicle to avoid a rear-end collision. The current study extends the CAMP FCW algorithm to improve the handling of extreme high-collision-risk scenarios. A total of 111 brake-only noncollision events was presented in the Tongji University Driving Simulator, and drivers' braking behaviors were used to model their ERDs. We found that ERDs depended on the interaction of LV deceleration and relative speed. In response to this finding, a nonlinear function with an interaction term was combined with a linear function into a piecewise function that accommodated both higher and lower LV deceleration conditions. The applicable domain of the warning onset range was then computed for a wide range of kinematic conditions. Results showed the piecewise function to be a better predictor of ERD than the linear function, as well as to result in fewer driver rejections of the FCWs.
Collision analysis often assumes emergency deceleration begins immediately upon completion of the vehicle’s mechanical brake lag. The goal of this study is to determine the driver-related delay from initial brake application to various degrees of deceleration in a simulated emergency and to test variables contributing to the modulation of driver braking. Using the data of Mazzae et al (2003), in which drivers respond to a lateral vehicle incursion, we have assessed the contribution of Time-to-Intersection (TTI), road condition, gender and crash outcome on driver emergency brake response. In the first 0.3 second phase after initial brake application, vehicle behavior was similar across all variables as drivers reached only moderate levels of deceleration. In the second phase, drivers often took more than one second to reach emergency decelerations, especially with a longer TTI. Pavement condition, gender and crash outcome were not significant factors. We discuss the consequences of driver braking behavior in the context of driver feedback and accident reconstruction analyses.
Meta-analyses have shown that the driver braking characteristics reported in the literature often vary due to differences in study methodologies (Green, 2000; Muttart, 2005; Summala, 2000). This paper provides additional insight on driver braking performance by 1) characterizing driver behavior during the execution of surprise and expected braking maneuvers, and 2) investigating the effect of gender, age, and vehicle driven on this behavior. Sixty-four drivers performed surprise and expected braking maneuvers from 45 mph (72.4 km/h) at an inflatable barricade, and subsequently performed expected braking maneuvers in response to an auditory alarm. Participants drove one of two instrumented vehicles: 1) a 2006 Mercedes-Benz R350, or 2) a 2007 Volvo S80. Drivers' braking inputs and corrected stopping distance were measured. Results indicate that drivers' braking performance varied by gender, age, and vehicle driven. These results and their relationship to the extant braking literature are discussed. The findings are expected to benefit those involved in crash reconstruction, researchers modeling driver braking behavior, as well as designers of brake-by-wire systems.
The present experiment was carried out to explore the response of driving subjects to emergency braking. The field trial consisted of driving behind a leading vehicle while the following drivers' responses were recorded by telemetry. A group of 51 individuals performed a series of trials at two driving speeds (60 and 80km/h), two following distances (6 and 12 m), and two braking conditions (real and dummy braking). Not all of these subjects completed all conditions or the minimum number of trials. The dependent variables were the total braking time (TBT) and its subcomponents: braking reaction time (BRT), and accelerator-to-brake movement time (MT). These data were analysed in three separate three-way ANOVAs with repeated measures on all factors. The results showed that when subjects were not aware of the forthcoming braking, the distance and braking conditions had major effects on all dependent variables. At the shorter following distance drivers reacted and moved faster. Similarly, when the brakes were real compared with dummy (i.e. brake lights only) drivers reacted faster. In addition, drivers reacted to onset of brake lights in 83% of the cases when dummy braking was applied, compared with 97% when real brakes were applied. Speed of driving did not show any significant effects and did not appear to influence the cognitive or attentional set to anticipate an emergency manoeuvre. These findings suggest that changes in angular velocity during optic expansion of the leading vehicle may be used as a cue to modulate braking movement, while onset of brake lights alone may be enough to trigger a ‘ballistic’ preventive response.
In his review on driver brake reaction times (RTs), Green (2000) rightly criticizes attempts to seek a canonical brake RT, and proposes to determine expected brake RT for specific situations. However, based on his analysis, he presents a series of values for expected, unexpected, and surprise situations that appear to generalize over a variety of different driver tasks and traffic situations without sufficient concern for urgency or criticality of the situations. This sampling problem may lead easily to biased and somewhat arbitrary estimates. Thus, instead of 1.25 sec for "unexpected" situations, the median yellow response time for the critical conditions (at short time-to-stop-line) is rather below 1.0 sec, and instead of 1.5 sec mean brake RT for surprise situations, available on-road data suggest that in fairly urgent situations-at time-to-collision of about 4.0 sec-unalerted drivers are able to react to an obstacle by braking at an average latency of 1.0 to 1.3 sec, depending on site. More emphasis should be given to analyzing (and producing) real-life data on driver reactions as a function of situational and driver-centered variables, and of criticality.
Time-headway (THW) during car-following and braking response were studied in a driving simulator from the perspective that behaviour on the manoeuvring level (e.g. choice of THW) may be linked to operational competence of vehicle control (e.g. braking) via a process of adaptation. Time-headway was consistent within drivers and constant over a range of speeds. Since time-headway represents the time available to the driver to reach the same level of deceleration as the lead vehicle in case it brakes, it was studied whether choice of time-headway was related to skills underlying braking performance. The initiation and control of braking were both affected by time-to-collision (TTC) at the moment the lead vehicle started to brake. This strongly supported the idea that time-to-collision information is used for judging the moment to start braking and in the control of braking. No evidence was found that short followers differ from long followers in the ability to accurately perceive TTC. There was, however, evidence that short followers are better able to programme the intensity of braking to required levels. Also, short followers tuned the control of braking better to the development of criticality in time during the braking process. It was concluded that short followers may differ from long followers in programming and execution of the braking response.
This study investigated whether drivers who operate a vehicle equipped with a front-to-rear-end collision warning system can avoid crashing when a lead vehicle brakes at its maximum potential (e.g., -0.85 g). Drivers in the experimental condition drove a 1993 General Motors Saturn mounted on the lowa Driving Simulator's six-degree-of-freedom motion base. The simulater cab was equipped with a collision warning vehicle and the lead vehicle. Two headway distances were tested (2.7 sec and 3.2 sec). The collision avoidance performance of subject drivers was compared to the behavior of drivers in a baseline condition where no collision warning display was present. Relative to the baseline condition, results indicate that drivers using the collision warning display (a) showed significantly fewer crashes in the shorter headway condition, (b) collided with the lead vehicle at significantly slower impact speeds, (c) released the accelerator significantly faster, and (d) had longer headways both at accelerator release and brake initiation.
To improve vehicle path-following performance and reduce driver workload, a human-centered feed-forward control (HCFC) system for a vehicle steering system is proposed. To be specific, a novel dynamic control strategy for steering ratio of vehicle steering systems that treats vehicle speed, lateral deviation, yaw error and steering angle as the inputs and a driver’s expected steering ratio as the output is developed. To determine the parameters of the proposed dynamic control strategy, drivers are classified into three types according to the
level of sensitivity to errors, i.e., low, middle and high level. The proposed HCFC system offers a human-centered steering system (HCSS) with a tunable steering gain, which can assist drivers in tracking a given path with smaller steering wheel angles and change rate of the angle by adaptively adjusting steering ratio according to driver’s path-following characteristics, reducing the driver workload. A series of experiments of tracking the centerline of double lane change (DLC) are conducted in Carsim and three different types of drivers are subsequently selected to test in a portable driving simulator under a fixed speed condition. The simulation and experiment results show that the proposed HCSS with the dynamic control strategy, compared to the classical control strategy of steering ratio, can improve task performance by about 7%, and reduce the driver’s physical workload and mental workload by about 35% and 50%, respectively, when following the given path.
Drivers’ last-second braking and last-second steering judgments have been studied extensively by the Crash Avoidance Metrics Partnership (CAMP) Forward Collision Warning (FCW) Requirements project. This previous work was conducted under closed-course conditions using a realistic surrogate target lead vehicle. In the current research, a subset of these tests involving more than 4000 individual test runs has been replicated in the National Advanced Driving Simulator (NADS) facility for comparison purposes.
The major conclusions from this research are as follows:
• Scenarios need to pay careful attention to ensure initial headway conditions prior to the critical approach event correspond to those that are typically experienced in real world driving. More generally, scenarios should have real-world validation.
• Scenarios should emphasize high lead vehicle decelerations. The 0.39-g deceleration levels gave the best results and have been used in previous CAMP surprise trial research.
• Scenarios should emphasize cases where the relative speed differential is high, particularly when the lead vehicle is stationary.
• Scenarios should emphasize last-second hard braking or hard steering over last-second “normal” maneuvers.
• Crash rates should not be used as a metric, and instead, attention should be focused on the interpretation of last-second maneuver onset behavior.
Investigations and analyses of braking behavior of inexperienced drivers in emergency situations have shown that the Brake Assist System which determines an appropriate emergency braking operation and assists in applying the necessary braking force is effective in terms of active safety. While the system designed to detect an emergency braking operation and to consistently apply the maximum braking force is effective for those drivers who cannot apply sufficient braking effort, the system will make experienced drivers feel disagreeable because the system may work against their intention to control the vehicle. Consequently, an ABS actuator was utilized to develop a type of Brake Assist System with excellent control performance and less disagreement. For the covering abstract see IRRD E102514.
Aims:
Myopia may have protective effects against diabetic retinopathy (DR). However, the data from epidemiologic studies are inconsistent. We aimed to examine the association between myopia and DR by conducting a meta-analysis.
Methods:
We identified studies by searching the PubMed and EMBASE databases. Study-specific odds ratios (ORs) were pooled using a fixed or random effects model. Myopic eyes were defined as having a spherical equivalent (SE)<-0.5 diopters (D). Myopic SE, each diopter decrease in SE toward myopia, and each millimeter increase in axial length (AL) were used as independent surrogate variables for myopia.
Results:
Data from 6 population-based and 3 clinic-based studies were included in the analyses. Myopic SE (compared with emmetropic eyes) and each millimeter increase in AL were associated with a decreased risk for DR (pooled odds ratio [OR], 0.80 and 0.79, respectively; 95% confidence interval [CI], 0.67-0.95 and 0.73-0.86, respectively; P=0.011 and 0.000, respectively). Each millimeter increase in AL was also associated with a decreased risk for vision-threatening diabetic retinopathy (VTDR) (pooled OR, 0.70; 95% CI, 0.60-0.82; P=0.000). No significant association between each diopter decrease in SE and DR was observed.
Conclusions:
Our meta-analysis suggests that individuals with myopia exhibit a decreased risk of developing DR or VTDR. An increased AL plays a critical role in this protective effect.
BAS assists driver's by automatically increasing their braking power during an emergency brake event when the driver is unable to apply a sufficient brake force. There are two performance requirements that BAS must fulfill in order to be employed effectively. One is the ability to activate when the driver suddenly applies brakes in an emergency while the other is the ability to provide additional assistance. Further study of BAS activation timing and degree of assistance in relation to driver acceptance is needed. The driver's acceptance of BAS refers to the BAS activation only during an emergency. A study was conducted to clarify drivers' emergency braking characteristics and measure the frequency of BAS activation during normal braking. One aim of the study was to verify driver characteristics during emergency braking on a test course. The study measured the brake pedal speed, force, and stroke during emergency braking along with the driver's compatibility with BAS activation conditions. Another task was to evaluate BAS with a driving simulator (DS). This study measured the frequency of BAS activation during normal braking by varying the BAS activation timing and degree of assistance. It also examined what the effects and side effects of varying these BAS parameters on the driver. In conclusion this study evaluates how varying BAS activation timing and its degree of assistance has on the drivers.
Dynamic Brake Support (DBS) is a safety system that has been applied to various passenger cars and has been shown to be effective at assisting drivers in avoiding or mitigating rear-end collisions. The objective of a DBS system is to ensure that the brake system is applied quickly and at sufficient pressure when a driver responds to a collision imminent situation. DBS is capable of improving braking response due to a passenger car driver's tendency to utilize multi-stage braking.Interest is developing in using DBS on commercial vehicles. In order to evaluate the possible improvement in safety that could be realized through the use of DBS, driver braking behavior must first be analyzed to confirm that improvement is possible and necessary. To determine if this is the case, a study of the response of truck drivers' braking behavior in collision imminent situations is conducted. This paper presents the method of evaluation and results.Data was drawn from a prior NHTSA simulator study and showed that many drivers exhibited multi-stage braking during four different imminent crash scenarios. Results from analysis indicate that a DBS system capable of detecting and intervening at the most conservative level of multi-staging presented could be successful in assisting drivers in avoiding or at minimum mitigating 30% of the collisions presented in this document. Further opportunities for assisting drivers based on reducing brake application time are also shown to exist. These results suggest that a DBS system may be valuable for heavy vehicles. .
Many experimental studies use repeated lead vehicle braking events to study the effects of forward collision warning (FCW) systems. It can, however, be argued that the use of repeated events induce expectancies and anticipatory behaviour that may undermine validity in terms of generalisability to real-world, naturalistic, emergency braking events. The main objective of the present study was to examine to what extent the effect of FCW on response performance is moderated by repeated exposure to a critical lead vehicle braking event. A further objective was to examine if these effects depended on event criticality, here defined as the available time headway when the lead vehicle starts to brake. A critical lead vehicle braking event was implemented in a moving-base simulator. The effects of FCW, repeated event exposure and initial time headway on driver response times and safety margins were examined. The results showed that the effect of FCW depended strongly on both repeated exposure and initial time headway. In particular, no effects of FCW were found for the first exposure, while strong effects occurred when the scenario was repeated. This was interpreted in terms of a switch from closed-loop responses triggered reactively by the situation, towards an open-loop strategy where subjects with FCW responded proactively directly to the warning. It was also found that initial time headway strongly determined response times in closed-loop conditions but not in open-loop conditions. These results raise a number of methodological issues pertaining to the design of experimental studies with the aim of evaluating the effects of active safety systems. In particular, the implementation of scenario exposure and criticality must be carefully considered.
While traffic accident fatalities in Japan have been declining, the number of injuries has continued on an upward trend for many years. One salient aspect of that rising trend is the number of casualties attributed to rear-end collisions. In 2005, such accidents accounted for approximately 35% of all fatalities and injuries. Regarding ordinary passenger cars, many of the drivers of the struck vehicles in rear-end collisions suffer slight neck injuries, while nearly all of the drivers of the striking vehicles are not injured. In this study, the influence of vehicle properties and human attributes on the incidence of neck injuries in rear-end collisions was analyzed using an integrated accident database developed by the Institute for Traffic Accident Research and Data Analysis (ITARDA). The results revealed, among other things, that an active head restraint system, which is one type of anti-whiplash device, is effective in suppressing the occurrence of neck injuries; that females tend to be injured more often than males; that age and generation influence the tendency for men to be injured; and that the trip purpose influences the tendency for neck injuries to occur. This tendency for generation and trip purpose to exert such an influence suggests the possibility that the health consciousness of the parties involved in rear-end collisions might affect the incidence of neck injuries. Among the other issues discussed in this paper is the concern that neck injuries due to rear-end collisions might increase in the future.
The design of active safety systems capable of helping avoiding a crash or reducing the collision severity requires data on how drivers behave in accident situations. These systems must be triggered when drivers actually need assistance. They must enhance insufficient reactions and limit unsuitable ones without being in conflict with drivers' natural behavior. The Laboratory of Accidentology, Biomechanics and human behavior, PSA Peugeot Citroën -Renault (LAB), has conducted experiments on driving simulators and on test tracks to analyze driver's behavior in emergency situations. Two of these experiments concern front-to-rear accident situations, each one involving more than 100 representative common drivers. The first study was carried out on a simulator with different accident scenarios : an adverse vehicle stopped or driving slowly at the top of a hill, a vehicle coming into the driver's lane from a parking area, or a vehicle driving in front of the subject then suddenly braking. The second study was carried out on a test track. The subjects were following a vehicle pulling a trailer that suddenly broke away and strongly braked. In both studies, the drivers' actions on the controls and the vehicle dynamics were recorded along with videos from driver's hands, feet and face and from the scene. The results show the benefit of such studies for the specification of active safety systems. These experiments revealed the inefficiency of braking actions of common drivers in emergency situations. The results provided a basis for the determination of triggering criteria of emergency brake assist and enabled to give recommendations on control strategies. Moreover, these experiments pointed out the benefit of emergency brake assist in terms of collision avoidance rate and crash speed reduction.
Driving is a task that requires the timely detection of critical events and relevant changes in traffic circumstances. Adaptation of speed and safety margins allows drivers to control the time available to react to potential hazards. One of the basic safety margins in driving is the time headway preserved with respect to cars ahead. To avoid rear-end collisions, drivers have to detect decelerations of lead cars. It can be assumed that fast or abrupt decelerations of the lead car are detected faster than slow or gradual decelerations. Moreover, expected decelerations are presumably detected faster than unexpected decelerations. Drivers' responses to rather abrupt and more gradual decelerations of the lead were investigated in a driving simulator. Situational traffic cues were used to manipulate the driver's expectations. Drivers adjusted the timing of their responses very well to the level of deceleration of the lead car. If cues in the environment indicated that the lead car was likely to decelerate, drivers reacted faster. Moreover, drivers increased their headway before the lead car actually started to decelerate, which can be considered an anticipatory response. In general, anticipation allows drivers to maintain their preferred headway and control time pressure in driving.
The minimum total braking time (i.e. the braking reaction time plus the accelerator-to-brake movement time) plays an important role in defining a minimum following gap (MFG). This study was designed to obtain a lower limit for this gap. Total braking times (TBT) of a group of 51 male and female young athletes were monitored during real driving conditions. Sudden braking applied by a leading private passenger vehicle initiated the trials. A within-subject design was used to study the effects of different factors on braking time. Individuals performed a series of semi-counterbalanced trials at two following distances (6 and 12 m), two speeds (60 and 80 km/h) and three expectancy stages (naïve driving, partial knowledge, and full knowledge of the forthcoming manoeuvre). A three-way repeated measures ANOVA showed no major effects of ‘speed’, but major effects of the ‘expectancy’ and the ‘distance’ factors. The experiment yielded a mean TBT of 0·678 s (SD = 0·144 s) for trials averaged over distances and speeds in the naïve condition only. The data emphasize the role played by pre-cues in the braking response prior to emergency stops. Both the level of awareness of the forthcoming manoeuvre and the distance between vehicles appear to determine the response time. The descriptive statistics presented may also provide the basis for an objective, acceptable and legally valid minimum time gap for prosecution of ‘careless’ drivers.
In many countries, motorcyclists are over-represented in traffic collision fatalities and injuries compared to vehicle registrations. Why drivers may violate the right-of-way of motorcyclists traveling as lead vehicles in front of drivers is empirically examined in two studies that were conducted with a moderate-fidelity driving simulator. The purpose of the first study was to determine if drivers, who also held a motorcycle license (N=16), drove cars differently than regular drivers (N=16) around motorcycles. The two groups did not differ on responses to motorcycling braking events, which was consistent with previous research on car following. The second study compared the driving performance of sixteen novice teenage drivers (M=16.2years of age) to 15 experienced drivers (M=32.9) over the span of six monthly simulator sessions. Novice drivers’ perception response times (PRT) to the braking events were significantly longer than those of the experienced drivers. PRTs to motorcycle and lead vehicle braking events decreased over sessions. For all participants, PRTs to the motorcycle events were longer than to the car events. The implications of these results for motorcyclists and drivers with different levels of experience are discussed.
Human perception-brake reaction time (RT) studies have reported a wide variety of results. By analyzing a large number of data sets, however, it is possible to estimate times under specific conditions. The most important variable is driver expectation, which affects RTs by a factor of 2. When fully aware of the time and location of the brake signal, drivers can detect a signal and move the foot from accelerator to brake pedal in about 0.70 to 0.75 sec. Response to unexpected, but common signals, such as a lead car's brake lights, is about 1.25 sec, whereas RTs for surprise events, such as an object suddenly moving into the driver's path, is roughly 1.5 sec. These times are modulated somewhat by other factors, including driver age and gender, cognitive load, and urgency.
The present paper describes a study concerned with the effect of alarm timing on driver trust and behaviour with a Forward Collision Warning System (FCWS). In this driving simulator experiment three different kinds of alarm timing (late/middle/early) were compared with respect to driver braking strategy and driver trust. The results showed that early alarm timing led to a more timely response to an imminent collision than either middle or late timing. With respect to driver trust, trust in late alarm timing is low compared with early or middle alarm timing. Furthermore the relationship between the timing of accelerator release and the timing of alarm onset is an important cause of changes in trust and if alarms are presented after drivers have already started to brake then trust is impaired. Middle alarm timing has the potential to improve driver braking response to imminent collisions. Possible benefits and drawbacks of FCWS are discussed from the viewpoint of alarm timing.
Forward collision warning (FCW) systems can reduce rear-end vehicle collisions. However, if the presentation of warnings is perceived as mistimed, trust in the system is diminished and drivers become less likely to respond appropriately. In this driving simulator investigation, 45 drivers experienced two FCW systems: a non-adaptive and an adaptive FCW that adjusted the timing of its alarms according to each individual driver’s reaction time. Whilst all drivers benefited in terms of improved safety from both FCW systems, non-aggressive drivers (low sensation seeking, long followers) did not display a preference to the adaptive FCW over its non-adaptive equivalent. Furthermore, there was little evidence to suggest that the non-aggressive drivers’ performance differed with either system. Benefits of the adaptive system were demonstrated for aggressive drivers (high sensation seeking, short followers). Even though both systems reduced their likelihood of a crash to a similar extent, the aggressive drivers rated each FCW more poorly than their non-aggressive contemporaries. However, this group, with their greater risk of involvement in rear-end collisions, reported a preference for the adaptive system as they found it less irritating and stress-inducing. Achieving greater acceptance and hence likely use of a real system is fundamental to good quality FCW design.
Warning timing and how drivers with and without forward collision warning (FCW) systems react when distracted at the moment a stationary vehicle is revealed directly ahead were investigated. The study was conducted using the Iowa Driving Simulator (IDS). The IDS was equipped with an FCW system that provided auditory warnings based on two warning criteria. A total of 30 subjects were split across three conditions - a baseline of 10 subjects (no warning display), and two warning conditions (early and late) with 10 subjects each. The two warning conditions differed by the duration of an a priori driver reaction component (1.5 and 1.0 s) in the warning algorithm. Drivers' collision avoidance performance in the two warning conditions was compared with that in the baseline condition. Results indicated that the early warning condition showed significantly shorter accelerator release reaction times, fewer crashes, and less severe crashes than both the baseline condition and the late warning condition. The results indicate that the timing of a warning is important in the design of collision warning systems.
Changes in drivers, vehicles, and roadways pose substantial challenges to the transportation safety community. Crash records and naturalistic driving data are useful for examining the influence of past or existing technology on drivers, and the associations between risk factors and crashes. However, they are limited because causation cannot be established and technology not yet installed in production vehicles cannot be assessed. Driving simulators have become an increasingly widespread tool to understand evolving and novel technologies. The ability to manipulate independent variables in a randomized, controlled setting also provides the added benefit of identifying causal links. This paper introduces a special issue on simulator-based safety studies. The special issue comprises 25 papers that demonstrate the use of driving simulators to address pressing transportation safety problems and includes topics as diverse as neurological dysfunction, work zone design, and driver distraction.
University of Michigan, Ann Arbor, ITS Research Center of Excellence http://deepblue.lib.umich.edu/bitstream/2027.42/1068/2/86649.0001.001.pdf
Perception of the lead car's braking was measured on-road when subjects of various levels of driving experience were looking at a digital display located at the lower part of the windscreen, at the speedometer level, or in the mid-console. The brake lights of the lead car were either working normally or switched off. The results indicated that the detection of the lead car's brake lights, in daylight, is substantially impaired when a following driver is looking at the speedometer area and brake lights do not contribute to detection at all when he/she is looking at a target in the mid-console. Driving experience did not influence performance in detecting a closing headway in peripheral vision, in contrast to improvement in lane-keeping found in an earlier study. It is suggested that such differential ability in using peripheral vision for lane and distance-keeping may mislead experienced drivers when they follow another vehicle and perform certain in-car tasks.
The total braking time (TBT) distribution is used as an input to two important traffic safety parameters: minimum following gap and stopping sight distance. It is therefore important to accurately estimate the TBT distribution. However, the previously published results on TBT distribution vary widely and confuse practitioners. In this paper, a meta-analysis is used in an effort to investigate the sources of variation in the studies of TBT. According to the results of the meta-analysis, significant characteristics of the mean of total braking time are the awareness level of the driver and the country in which the experiment took place. In addition to these two characteristics, both the type of brake stimulus and the distance away from the brake stimulus are found to be influential characteristics on the variance of total braking time. Based on several combinations of these factors, TBT distributions are reconstructed. It is recommended that the percentile estimates of TBT used for the minimum following gap and stopping sight distance need to be adjusted.
Rear-end collisions account for almost 30% of automotive crashes. Rear-end collision avoidance systems (RECASs) may offer a promising approach to help drivers avoid these crashes. Two experiments performed using a high-fidelity motion-based driving simulator examined driver responses to evaluate the efficacy of a RECAS. The first experiment showed that early warnings helped distracted drivers react more quickly--and thereby avoid more collisions--than did late warnings or no warnings. Compared with the no-warning condition, an early RECAS warning reduced the number of collisions by 80.7%. Assuming collision severity is proportional to kinetic energy, the early warning reduced collision severity by 96.5%. In contrast, the late warning reduced collisions by 50.0% and the corresponding severity by 87.5%. The second experiment showed that RECAS benefits even undistracted drivers. Analysis of the braking process showed that warnings provide a potential safety benefit by reducing the time required for drivers to release the accelerator. Warnings do not, however, speed application of the brake, increase maximum deceleration, or affect mean deceleration. These results provide the basis for a computational model of driver performance that was used to extrapolate the findings and identify the most promising parameter settings. Potential applications of these results include methods for evaluating collision warning systems, algorithm design guidance, and driver performance model input.
In order to improve road safety, automobile manufacturers are now developing Forward Collision Warning Systems (FCWS). However, there has been insufficient consideration of how drivers may respond to FCWS. This driving simulator study focused on alarm timing and its impact on driver response to alarm. The experimental investigation considered driver perception of alarm timings and its influence on trust at three driving speeds (40, 60 and 70 mile/h) and two time headways (1.7 and 2.2 s). The results showed that alarm effectiveness varied in response to driving conditions. Alarm promptness had a greater influence on ratings of trust than improvements in braking performance enabled by the alarm system. Moreover, alarms which were presented after braking actions had been initiated were viewed as late alarms. It is concluded that drivers typically expect alarms to be presented before they initiate braking actions and when this does not happen driver trust in the system is substantially decreased.
Rear-end collisions are often quoted as being a major cause of road traffic accidents. In response to this, a great deal of ergonomics research effort has been directed towards the analysis of brake reaction times. However, the engineering solution has been to develop advanced systems for longitudinal control, which, it is argued, will mitigate the problem of rear-end collisions. So far, though, there have been few empirical studies to determine how brake reaction times will be affected by such vehicle automation. This paper presents a literature review summarizing the current state of knowledge about driver responses in non-automated vehicles. The review covers driver factors, vehicle factors and situational factors. Following the review, some empirical data are presented from a driving simulator experiment assessing brake reaction times of skilled and unskilled drivers under two different levels of automation. When compared to previous data gathered during manual driving, there seems to be a striking increase in reaction times for these automated conditions. Implications for the design and safety of automated vehicle systems are discussed.
Scenario criticality determines the effects of working memory load on brake response time
- Engström
Engström, J., 2010. Scenario criticality determines the effects of working memory load on brake response time. In: Krems, J., Petzoldt, T., Henning, M. (Eds.),
Proceedings of the European Conference on Human Centered Design for Intelligent Transport Systems (HUMANIST), Lyon, France, pp. 25-36.
A study on driver behavior during braking on open road
- M Kassaagi
- G Brissart
- J C Popieul
Kassaagi, M., Brissart, G., Popieul, J.C., 2003. A study on driver behavior during braking on open road. In: 18th International Technical Conference on the
Enhanced Safety of Vehicles (ESV), Nagoya (Japan), May.
Development and Validation of Functional Definitions and Evaluation Procedures for Collision Warning/Avoidance Systems. Publication DOT HS 808 964
- R J Kiefer
- D J Leblanc
- M Palmer
- J Salinger
- R Deering
- M Shulman
Kiefer, R.J., LeBlanc, D.J., Palmer, M., Salinger, J., Deering, R., Shulman, M., 1999. Development and Validation of Functional Definitions and Evaluation
Procedures for Collision Warning/Avoidance Systems. Publication DOT HS 808 964. NHTSA, U.S. DOT.
The accident avoidance potential of the motor vehicle: accident data, vehicle handling and safety standards
- Limpert
Limpert, R., Gamero, F.E., 1974. The accident avoidance potential of the motor vehicle: accident data, vehicle handling and safety standards. Proceedings of
the Third International Congress on Automotive Safety, vol. 11.
NHTSA Light Vehicle Antilock Brake Systems Research Program Task 5.2/5.3: Test Track Examination of Drivers' Collision Avoidance Behavior Using Conventional and Antilock Brakes
- E Mazzae
- F Barickman
- G Forkenbrock
- G Baldwin
Mazzae, E., Barickman, F., Forkenbrock, G., Baldwin, G., 2003. NHTSA Light Vehicle Antilock Brake Systems Research Program Task 5.2/5.3: Test Track
Examination of Drivers' Collision Avoidance Behavior Using Conventional and Antilock Brakes. National Highway Transportation Safety Administration,
Washington, DC.
Estimating Driver Response Times
- J W Muttart
Muttart, J.W., 2005. Estimating Driver Response Times.
Publication DOT HS 812 139. NHTSA
Traffic Safety Facts 2013, 2013. Publication DOT HS 812 139. NHTSA, U.S. DOT.
Shanghai 2020 Driving Scenario Models and Traffic Accident Models Development
- X Wang
- X Chen
- B Deng
Wang, X., Chen, X., Deng, B., 2011. Shanghai 2020 Driving Scenario Models and Traffic Accident Models Development. General Motor, Shanghai, Unpublished
Report.
A study on driver behavior during braking on open road
- Kassaagi
Development of a kinematic-based forward collision warning algorithm using an advanced driving simulator
- Wang