Figure - available from: Journal of Advanced Transportation
This content is subject to copyright. Terms and conditions apply.
Source publication
In this study, we selected four real-world rear-end crash scenarios with different crash characteristics. The vehicles involved in those crashes were not equipped with any crash avoidance systems. We then used the accident reconstruction method to build those crash scenarios in PC-Crash software. Then, different FCW/AEB safety algorithms have been...
Citations
... Therefore, professional traffic crash simulation software, PC-Crash can be used to generate crash trajectory data. The software is capable of producing the trajectory of the vehicle both before and after the crash [42,43]. To ensure similarity between simulated and authentic data, the vehicle's acceleration, speed, maximum acceleration and steering angle are adjusted randomly within the given range. ...
Efficient detection of traffic crashes has been a significant matter of concern with regards to expressway safety management. The current challenge is that, despite collecting vast amounts of data, expressway detection equipment is plagued by low data utilization rates, unreliable crash detection models, and inadequate real‐time updating capabilities. This study is to develop an effective digital twin framework for the detection of traffic crashes on expressways. Firstly, the digital twin technology is used to create a virtual entity of the real expressway. A fusion method for macro and micro traffic data is proposed based on the location of multi‐source detectors on a digital twin platform. Then, a traffic crash detection model is developed using the ThunderGBM algorithm and interpreted by the SHAP method. Furthermore, a distributed strategy for detecting traffic crashes is suggested, where various models are employed concurrently on the digital twin platform to enhance the general detection ability and reliability of the models. Finally, the efficacy of the digital twin framework is confirmed through a case study of certain sections of the Nanjing Ring expressway. This study is expected to lay the groundwork for expressway digital twin studies and offer technical assistance for expressway traffic management.
... On the basis of the relative speeds of the vehicles, it is possible to convert the DTC to the TTC. 35 We calculated the distance-to-collision (DTC), which is the distance between the frontmost point of the rider's front wheel and the rearmost point of the obstacle's rear wheel at the beginning of the avoidance maneuver. ...
As vehicle applications have evolved to a more intelligent and self-driving stage, autonomous vehicles have attracted more attention in recent years. This paper proposes a trajectory planner that considers feasibility, safety and passenger acceptance. This will ensure autonomous vehicles satisfy the constraints of the traffic environment, driving ability, and comfort drivers experience during collision avoidance. This paper deals with planning collision-free trajectories for autonomous vehicles on highways. The problem is formulated using reachability-based planning via zonotope. According to the vehicle dynamics model, the trajectory feasibility is determined by the vehicle motion feasibility set. The next step is to apply safety constraints to the base planner by collision avoidance checking. Given that this planner uses a receding horizon strategy, it selects a safe parameter in each planning iteration. At each stage of planning, the set of reachable vehicles should not intersect with any obstacles. Since braking cannot prevent a collision, this approach consists of lane changing and overtaking maneuvers to avoid collisions. Finally, knowledge from the safety of the intended functionality (SOTIF) standard is utilized to verify the algorithm performance. The efficiency and performance of different driving styles of trajectory planners are verified by vehicle tests under different vehicle velocities and different obstacle disturbances. Satisfactory results are obtained from the set of simulated scenarios.
... When using a 13 × 13 window size, the number one cell is generated using thirty-six pixel values from (1, 1) coordinate to (6,6) coordinate, as shown in (4). In the number two cell, it is generated by using thirty-six pixel values from (1,8) coordinate to (6,13) coordinate, as shown in (5). In the number three cell, it is generated by using thirtysix pixel values from (8, 1) coordinate to (13,6) coordinate, as shown in (6). ...
Embedded stereo vision systems based on traditional approaches often require a disparity refinement process to enhance image quality. Weighted median filter (WMF)-based processors are commonly employed for their excellent refinement performance. However, when implemented on a field-programmable gate array (FPGA), WMF-based processors face a trade-off between hardware resource utilization and refinement performance. To address this trade-off, we previously proposed a new disparity refinement processor based on the hybrid max-median filter (HMMF). However, our earlier work did not guarantee flawless operation in large occluded and texture-less regions, particularly in areas with numerous holes. In order to overcome this limitation of conventional processors, we proposed a cell-based disparity refinement processor. This processor extends our previous HMMF-based disparity refinement processor. To evaluate its refinement performance, we conducted experiments using four types of publicly available stereo datasets. When comparing refinement performance, our proposed processor outperforms conventional processors when using the KITTI 2012 and 2015 stereo benchmark datasets. Additionally, the results demonstrate that our proposed processor exhibits superior refinement performance when applied to the Cityscapes and StereoDriving datasets in comparison to conventional processors. Furthermore, when considering hardware resource utilization, our proposed processor demonstrates lower resource requirements than conventional processors when implemented on an FPGA. Therefore, our proposed disparity refinement processor is well-suited for the disparity refinement process in stereo vision systems that require cost-effectiveness and high performance.
... Forward Collision Warning represents an advanced technology to help drivers avoid collisions, particularly rear-end collisions when both vehicles are traveling in the same direction (see Fig. 1), thus addressing impaired ability of drivers to react quickly e.g., to emergency braking of vehicles in front (Bella & Russo, 2011;Kusano & Gabler, 2011;Seaman et al., 2022;Seyedi et al., 2021). ...
... FCW is typically (as is RECAS) implemented as part of a collision avoidance system (CAS) or collision warning system (CWS), which utilizes radar, lidar and computer vision systems to identify a high probability of collision (Bella & Russo, 2011;Kusano & Gabler, 2011 American Automobile Association, 2019, as cited in Seaman et al., 2022;Seyedi et al., 2021). The FCW system warns the driver of the possibility of a collision when TTC becomes critically short (Kusano & Gabler, 2011Seyedi et al., 2021). ...
... FCW is typically (as is RECAS) implemented as part of a collision avoidance system (CAS) or collision warning system (CWS), which utilizes radar, lidar and computer vision systems to identify a high probability of collision (Bella & Russo, 2011;Kusano & Gabler, 2011 American Automobile Association, 2019, as cited in Seaman et al., 2022;Seyedi et al., 2021). The FCW system warns the driver of the possibility of a collision when TTC becomes critically short (Kusano & Gabler, 2011Seyedi et al., 2021). The operation principle of FCW is therefore similar to RECAS, however while RECAS is installed in the LV, FCW is installed in the FV. ...
Every year, several thousand powered two-wheeler (PTW) i.e., motorcycle, moped and scooter drivers and passengers die in traffic accidents in the EU. Despite the much higher risk of death and injuries for PTW vs. car users, there is a threefold lack regarding collision warning technologies for PTWs: lack of research, lack of regulation and lack of availability in the market. Many injuries occur in rear-end collisions, when PTW is struck from the rear by another vehicle. In this paper we present a hybrid, multi-method simulation model that allows simulation of various situations in which a vehicle may collide with the rear end of a PTW. We have used this model to estimate the potential impact of market penetration of a novel PTW ESS + RECAS system, named MEBWS (Motorcycle Emergency Braking Warning System), within the EU on the number of traffic accidents and their consequences, which would contribute to the EU "Vision Zero" goal: "reduce road deaths to almost zero by 2050". MEBWS has been developed at the Faculty of Information Studies in Novo mesto and patented. Simulation results using EU traffic accident data show that with 100% market penetration of the MEBWS system in the EU, the total number of PTW rear-end collisions would decrease by 29.50%. This reduction would result in fewer injuries and a decrease in economic crash costs by €43,145,172, according to the standard EU methodology. With the MEBWS system enabled, the number of traffic accidents in the standard rear-end collision emergency braking scenarios Moto, normal drive, Moto, emergency stop and Moto, not moving decreased by 33.15%, 27.76% and 28.76%, respectively. In cases where the collision could not be prevented due to slow response of the following driver or very high relative speed of the vehicles, MEBWS reduced the relative speed at impact, resulting in a reduction of injury severity by up to 11.198%, as estimated by the amount of kinetic energy released at collision.
... We mark them on the indices of the RTE and MiD. The threshold for crucial cases is determined by rounding the typical TTC threshold of 2.7 seconds used in FCW systems [33,44]. This division allows a more detailed analysis of the performance for the TTC estimation algorithms in different TTC intervals, providing a better understanding of their limitations and strengths. ...
Time-to-Contact (TTC) estimation is a critical task for assessing collision risk and is widely used in various driver assistance and autonomous driving systems. The past few decades have witnessed development of related theories and algorithms. The prevalent learning-based methods call for a large-scale TTC dataset in real-world scenarios. In this work, we present a large-scale object oriented TTC dataset in the driving scene for promoting the TTC estimation by a monocular camera. To collect valuable samples and make data with different TTC values relatively balanced, we go through thousands of hours of driving data and select over 200K sequences with a preset data distribution. To augment the quantity of small TTC cases, we also generate clips using the latest Neural rendering methods. Additionally, we provide several simple yet effective TTC estimation baselines and evaluate them extensively on the proposed dataset to demonstrate their effectiveness. The proposed dataset is publicly available at https://open-dataset.tusen.ai/TSTTC.
... Countermeasures, such as advanced driver assistance systems, are constantly developed to mitigate these risks. One way to evaluate such systems before they enter the market is through virtual simulations (Anderson et al., 2013;Seyedi et al., 2021). Since such evaluations are performed in a virtual rather than physical test environment, they are more cost-efficient than traditional test beds. ...
Subsampling is commonly used to overcome computational and economical bottlenecks in the analysis of finite populations and massive datasets. Existing methods are often limited in scope and use optimality criteria (e.g., A-optimality) with well-known deficiencies, such as lack of invariance to the measurement-scale of the data and parameterisation of the model. A unified theory of optimal subsampling design is still lacking. We present a theory of optimal design for general data subsampling problems, including finite population inference, parametric density estimation, and regression modelling. Our theory encompasses and generalises most existing methods in the field of optimal subdata selection based on unequal probability sampling and inverse probability weighting. We derive optimality conditions for a general class of optimality criteria, and present corresponding algorithms for finding optimal sampling schemes under Poisson and multinomial sampling designs. We present a novel class of transformation- and parameterisation-invariant linear optimality criteria which enjoy the best of two worlds: the computational tractability of A-optimality and invariance properties similar to D-optimality. The methodology is illustrated on an application in the traffic safety domain. In our experiments, the proposed invariant linear optimality criteria achieve 92-99% D-efficiency with 90-95% lower computational demand. In contrast, the A-optimality criterion has only 46% and 60% D-efficiency on two of the examples.
... Otherwise, the current driving state is safe by default. In addition, Seyedi M. et al. [16] selected four real-world rear-end collision scenarios with different collision characteristics, defined different forward collision warning (FCW)/AEB safety algorithms, and simulated important input parameters that affect collision results, such as speed, braking strength and driver reaction time. Bae J.J. et al. [17] designed a hierarchical braking control strategy based on collision time, and developed an autonomous braking algorithm that can satisfy both vehicle safety and riding comfort, which consists of a two-step braking strategy dependent on collision time. ...
In order to improve the active safety of vehicles, the control strategy of the vehicle automatic emergency braking system is studied. The hierarchical control idea is used to model the control strategy. The upper controller is a collision time model for the decision-making of vehicle braking deceleration, and the collision time threshold is determined under the condition of considering comfort. According to the braking deceleration output by the upper controller, the lower controller controls the vehicle by adjusting the throttle opening and braking pipeline pressure through PID control. Based on the typical test conditions of C-NCAP, a joint simulation test of CarSim and Matlab/Simulink for hierarchical control strategy is carried out. In order to achieve further verification, several groups of test conditions are conducted, and finally its effectiveness is verified, which can ensure the safety of drivers.
... Intersection collision avoidance technologies including, but not limited to, automated traffic signal performance measure (45), cooperative intersection collision avoidance system for violations (46), and autonomous emergency brake on forward collision warning systems (47) can be helpful. These infrastructure-to-vehicle-and vehicle-to-vehicle communication systems are able to improve the safety condition of the locations prone to this hazard particularly during apparent sunrise and sunset. ...
There has been extensive visual-disability-focused safety research resulting from various factors/conditions that have a significant impact on a driver’s ability to maneuver safely. One such critical condition among the major environmental issues that contribute to traffic crashes is daytime sun glare-induced temporary visual impairment. Nevertheless, the contribution of relevant factors to the choice of potential actions at the time of such crashes has not been widely studied. Approximately 3,000 crashes per year in the United States are attributed to sun glare, which may indicate that the driver had no control over the situation. Thus, this study investigated several factors to identify crashes that occurred, with acceptable certainty, under the direct influence of sun glare over 4 years (2014 to 2017) in Florida—“the Sunshine State.” A multinomial logistic regression model was developed to formulate relationships between crash-related contributing factors and to evaluate the likelihood of alternative actions being adopted by drivers experiencing sun glare. The results indicated that running red lights/stop signs was the most likely action one might take while experiencing this adverse condition, particularly on local roadways. A further consequence of sun glare is that drivers tend to follow the car in front too closely along segments with higher annual average daily traffic. The findings could help safety officials support the use of emerging intelligent transportation system technologies (i.e., automated traffic signal performance measures and cooperative intersection collision avoidance systems) to alert drivers of the possibility of a temporary vision impairment exactly when high-glare conditions exist.
... However, when developing both advanced driver assistance systems (ADAS) and autonomous driving systems (ADS), there is a need to assess the impact on safety of the systems before they are on the market. One way to do that is by running virtual simulations comparing the outcome of simulations both with and without a specific system (Anderson et al., 2013;Sander, 2018;Seyedi et al., 2021). Such simulations are often called counterfactual simulations, as they assess what could have happened if the system were on the road. ...
Data subsampling has become widely recognized as a tool to overcome computational and economic bottlenecks in analyzing massive datasets and measurement-constrained experiments. However, traditional subsampling methods often suffer from the lack of information available at the design stage. We propose an active sampling strategy that iterates between estimation and data collection with optimal subsamples, guided by machine learning predictions on yet unseen data. The method is illustrated on virtual simulation-based safety assessment of advanced driver assistance systems. Substantial performance improvements were observed compared to traditional sampling methods.
Jaywalking of electric bicycles (e-bikes) is dangerous because it can catch drivers of motor vehicles off guard and lead to collisions between e-bikes and motor vehicles. Hence, this study set out to investigate e-bike rider-injury severity in collisions between jaywalking e-bikes and cars. Based on large-scale simulation experiments using PC-Crash, the effects of collision angle, relative position, and car speed on rider-injury severity were examined. Dynamic response of e-bike riders’ head, chest, and lower extremity at different influencing factor levels were compared using boxplots and a one-way analysis of variance. The least significant difference test was further performed for posthoc pairwise comparisons. The results showed that crossing the road perpendicular to the car direction is the safest crossing angle for e-bike riders. Riders are less injured when hitting cars’ left front corner. As car speed increases, e-bike riders’ head and chest injuries become more severe.
