Car-following Safety Research Articles

An Efficient Safety-Oriented Car-Following Model for Connected Automated Vehicles Considering Discrete Signals

With the rapid development of Connected and Automated Vehicle (CAV) technology, limited self-driving vehicles have been commercially available in certain leading intelligent transportation system countries. When formulating the car-following model for CAVs, safety is usually the basic constraint. Safety-oriented car-following models seek to specify a safe following distance that can guarantee safety if the preceding vehicle were to brake hard suddenly. The discrete signals of CAVs bring a series of phenomena, including discrete decision-making, phase difference, and discretely distributed communication delay. The influences of these phenomena on the car-following safety of CAVs are rarely considered in the literature. This paper proposes an efficient safety-oriented car-following model for CAVs considering the impact of discrete signals. The safety constraints during both normal driving and a sudden hard brake are incorporated into one integrated model to eliminate possible collisions during the whole driving process. The mechanical delay information of the preceding vehicle is used to improve car-following efficiency. Four modules are designed to enhance driving comfort and string stability in case of heavy packet losses. Simulations of a platoon with diversified vehicle types demonstrate the safety, efficiency, and string stability of the proposed model. Tests with different packet loss rates imply that the model could guarantee safety and driving comfort in even poor communication environments.

Effect of Five Driver's Behavior Characteristics on Car-Following Safety.

Driver's behavior characteristics (DBCs) influence car-following safety. Therefore, this paper aimed to analyze the effect of different DBCs on the car-following safety based on the desired safety margin (DSM) car-following model, which includes five DBC parameters. Based on the Monte Carlo simulation method, the effect of DBCs on car-following safety is investigated under a given rear-end collision (RECs) condition. We find that larger subjective risk perception levels can reduce RECs, a smaller acceleration sensitivity (or a larger deceleration sensitivity) can improve car-following safety, and a faster reaction ability of the driver can avoid RECs in the car-following process. It implies that DBCs would cause a traffic wave in the car-following process. Therefore, a reasonable value of DBCs can enhance traffic flow stability, and a traffic control strategy can improve car-following safety by using the adjustment of DBCs.

A general approach to smoothing nonlinear mixed traffic via control of autonomous vehicles

Stop-and-go waves are easily caused by unstable traffic due to the collective behavior of human drivers, resulting in higher vehicle fuel consumption and emissions. In this article, we develop a general approach to synthesizing feedback controllers of autonomous vehicles (AVs) for smoothing unstable nonlinear mixed traffic flow, consisting of AVs and human-driven vehicles (HVs). In the context of traffic smoothing and stabilization, prior studies have mainly focused on analyzing string stability of specific car-following models linearized at flow equilibrium points. By contrast, we are interested to analytically synthesize appropriate feedback controllers of AVs for smoothing nonlinear mixed traffic in its general functional forms, covering a broad class of deterministic car-following models commonly seen in the literature. Specifically, by leveraging feedback control theory AVs are controlled to operate in such a way that they closely track a virtual speed profile, i.e., a subtler version of the disturbance resulting from the immediate preceding vehicle. Thus, traffic waves are reduced when propagating backward across controlled AVs. Based on the general functional form of car-following dynamics, we derive a class of effective additive AV controllers that are proven to be able to ensure convergence in tracking desired speed profiles, leading to smoother traffic. In addition, a set of sufficient conditions is devised for guaranteeing car-following safety. More importantly, head-to-tail string stability of the mixed traffic is likely to be achieved with a sufficient market penetration rate (MPR) of AVs. Notably, unlike many existing studies the feedback controllers synthesized for AVs require only local traffic information without having to rely on high degrees of vehicle connectivity, and the rate of smoothing mixed traffic is readily tunable, which is useful for practical implementations. The proposed approach is further illustrated with a theoretical intelligent driver model (IDM) and commercially available adaptive cruise control (ACC) vehicles represented by a well calibrated IDM. Extensive numerical results are presented to show the effectiveness and robustness of the feedback controllers synthesized for AVs on smoothing nonlinear mixed traffic.

Developing an improved automatic preventive braking system based on safety-critical car-following events from naturalistic driving study data

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.

Effects of assignments of dedicated automated vehicle lanes and inter-vehicle distances of automated vehicle platoons on car-following performance of nearby manual vehicle drivers

Deploying dedicated lanes for automated vehicles (AVs) can effectively alleviate the coordination issues between AVs and manual vehicles (MVs). However, AV platoons running on dedicated AV lanes (DAVLs) have a prominent collective behavior characteristic of small inter-vehicle distance. The nearby MV drivers’ imitation of this characteristic may reduce their car-following time headway (THW). The researchers conducted a simulation experiment to investigate the influence of DAVL assignments, inter-vehicle distances of AV platoons and AV platoon speed on the car-following performance of nearby MV drivers. The data of mean THW, standard deviation of THW, standard deviation of lateral position, standard deviation of velocity, standard deviation of horizontal gaze position and mean saccadic peak velocity were collected from 36 participants. Statistical analysis results show that the three factors considerably affected the MV drivers’ car-following performance. In particular, the MV drivers showed a worse car-following safety but a better driving stability when the left lane was dedicated to AVs than when the right lane was dedicated to AVs (Note the experiments were done in a drive-on-the-left environment.). With respect to the inter-vehicle distances of AV platoons, the MV drivers’ car-following safety was poorer under the 4 m condition than that under the 10 and 18 m conditions. In addition, the MV drivers showed a poorer car-following safety and bore a larger mental workload when driving next to the AV platoons running at 110 km/h. This study may provide some suggestions for DAVLs. Assigning the right lane of a three-lane motorway as the DAVL may have a slighter negative impact on the nearby MV drivers in China. In terms of traffic management in DAVLs, the inter-vehicle distance of AV platoons can be reduced to 10 m, and the speed of AVs should not be higher than the design speed of adjacent MV lanes.

Modeling Analysis of Improved Minimum Safe Following Distance under Internet of Vehicles

The development of the Internet of vehicles technology can improve the communication between vehicles, thereby changing the driving behavior of drivers. Therefore, the traditional safe-following model cannot accurately describe the driving behavior and needs to be improved accordingly. First, two key parameters (i.e., drivers’ reaction sensitivity and road friction coefficient) are obtained through a comprehensive comparative analysis of influencing factors on the Internet of vehicles environment. And the calculation methods of these two parameters are proposed by using the multilevel comprehensive weighted evaluation method and the BP neural network. Then, these two key parameters are used to improve the traditional minimum safety distance model for adapting to driving behavior under the Internet of vehicles environment. Finally, through setting up simulation experiments and comparative analysis, the relationship between different influencing factors and the minimum safe following distance is obtained, and the influence degree of different influencing factors is sorted. The most important factor affecting car-following safety is the drivers’ characteristics. It can provide strong theoretical support for the safe driving assistance system of vehicles.

A matched case-control method to model car-following safety

Car-following safety is related to both observed driving characteristics (e.g. car-following behaviour) and unobserved driver heterogeneity (e.g. drivers’ psychological features). Two major issues remain in the existing literature, i.e. limiting to longitudinal characteristics and not addressing the confounding effects of unobserved driver heterogeneity. This study takes a matched case–control approach to model car-following safety with both longitudinal and lateral driving characteristics. Unobserved driver heterogeneity is controlled by matching preceding and following vehicle IDs. Results show that unstable lateral movements of preceding vehicles and following vehicles contribute to higher crash risks. Comparison results on two datasets with different congestion levels reveal that it is safer in more congested traffic when the following vehicle maintains more stable longitudinal and lateral behaviours, and greater speed difference, headway, and spacing regarding its preceding vehicle. This study provides insights in enhancing roadway safety management and benefiting the automated vehicle development by warnings on associated risks.

An empirical study on heterogeneous traffic car-following safety indicators considering vehicle types

Considering different car-following characteristics of trucks and cars, this study verifies the heterogeneity of steady car-following behavior of four leader-follower vehicle-type combinations based on the naturalistic driving dataset. The best-fit distribution model of surrogate safety indicators (i.e., distance headway (DHW), time headway (THW), time to collision (TTC) and safety margin (SM)) of four combinations are identified. Then, the performance of surrogate safety indicators in various speed ranges is compared to select a homeostatic risk perception indicator. After that, the longitudinal control model is calibrated for four combinations. Results show that the heterogeneity of different leader-follower vehicle-type combinations has been confirmed by surrogate safety indicators and calibrated parameters of the longitudinal control model. SM is suitable as the homeostatic risk perception indicator for four combinations when the speed is less than 90 km/h, and the advised safety warning SM value of is 0.9 or 0.95 for four combinations.

Operational safety of automated and human driving in mixed traffic environments: A perspective of car-following behavior

The advent of automated vehicles (AVs) will provide opportunities for safer, smoother, and smarter road transportation. During the transition from the current human-driven vehicle (HV) to a fully AV traffic environment, there will be a mixed traffic flow including both HVs and AVs. The impact of introducing AVs into existing traffic, however, has not yet been fully understood. In this paper, we advance this understanding by conducting mixed traffic safety evaluation from the perspective of car-following behavior using real-world AV operational data of mixed traffic. To understand how the AVs impact other vehicles on the road, we analyzed the operational behaviors of HV-following-HV, AV-following-HV, and HV-following-AV. A selected car-following model is calibrated, and results show that there are significant differences between the HV-following-HV and the other two groups, indicating safe AV behavior and changes in HV behavior (i.e. less aggressive, safer) after the introduction of AVs into the traffic. Additionally, to understand AV behavioral safety, we investigate behavior predictions (one of the most critical inputs for AVs to make car-following decisions) of AVs and their surrounding vehicles using a mature baseline model and a new Conditional Variational Autoencoder (CVAE) framework. The result shows potential risks of inaccurate predictions of the baseline model and the necessity to consider additional factors, such as vehicle interactions and driver behavior, into the prediction for risk mitigation. Arterial vehicle trajectory data from the Lyft Level 5 Dataset is applied to test the proposed methodological framework to understand the car-following safety risks of HVs and AVs in the mixed traffic stream.

String Stability and Platoon Safety Analysis of a New Car-Following Model Considering a Stabilization Strategy

Adaptive cruise control (ACC) systems can reduce collision risk and make traffic flow more smoothly; nevertheless, improving the string stability and car-following safety in ACC systems has remained an important research topic. Based on the desired safety margin (DSM) model adopted as an ACC velocity control method, a sliding mode controller is proposed to investigate string stability and car-following safety using the time headway policy (THP), and its stability is verified by the Lyapunov stability theory. Furthermore, numerical simulations are conducted to verify the effectiveness of the proposed stabilization strategy for the stability of the DSM model. Analyzing the risk assessment indexes (time-to-collision, TTC, and time headway, TH) of the DSM model reveals that the proposed stabilization strategy can improve the traffic flow stability and avoid rear-end collision risks when the leading car exhibits a small disturbance. Therefore, the proposed stabilization strategy is valuable for designing ACC controllers to enhance traffic flow stability and car-following safety in automotive platoon driving.

Effect of the Uncertainty Level of Vehicle-Position Information on the Stability and Safety of the Car-Following Process

In recent years, the adaptive cruise control (ACC) system has become widely adopted. The vehicle-positioning system plays an important role in automotive platoon driving and can provide vehicle-position information through vehicle-to-vehicle communication. However, the acquired vehicle-position information often presents uncertainties given the existence of variations in the traffic environment and of noise in the vehicle-positioning system. These uncertainties, in turn, influence the car-following performance of a vehicle platoon. In this study, we employ the desired safety margin (DSM) model, an ACC control strategy, to investigate the influence of the uncertainty level of vehicle-position information (ULVPI) on string stability and car-following safety. The stability criterion of the DSM model with ULVPI is derived through linear stability theory. Theoretical analysis results are verified by numerical simulations. Analytical results indicate that a negative ULVPI value can expand the stable region and improve string stability and that a positive ULVPI value can reduce delay time. Moreover, a negative ULVPI value can improve car-following safety during the stopping and evolution processes, whereas a positive ULVPI values can increase the safety margins of vehicles during the starting process. In the car-following process, a negative ULVPI value can improve car-following safety and reduce rear-end collision risk. The variation in ULVPI values can improve string stability and reduce the risk of rear-end collisions when mean and standard deviation are reasonable. Therefore, the improvement of a vehicle’s dynamic performance, string stability, and safety relies on the effect of ULVPI in different traffic scenarios. These results are useful in designing different control strategies that stabilize traffic flow and improve traffic safety for vehicles with ACC systems.

Study and Simulation Analysis of Vehicle Rear-End Collision Model considering Driver Types

The reasonable distance between adjacent cars is very crucial for roadway traffic safety. For different types of drivers or different driving environments, the required safety distance is different. However, most of the existing rear-end collision models do not fully consider the subjective factor such as the driver. Firstly, the factors affecting driving drivers’ characteristics, such as driver age, gender, and driving experience are analyzed. Then, on the basis of this, drivers are classified according to reaction time. Secondly, three main factors affecting driving safety are analyzed by using fuzzy theory, and the new calculation method of the reaction time is obtained. Finally, the improved car-following safety model is established based on different reaction time. The experimental results have shown that our proposed model obtained more accurate vehicle safety distance with varied traffic kinematic conditions (i.e., different traffic states, varied driver types, etc.). The findings can help traffic regulation departments issue early warnings to avoid potential traffic accidents on roads.

Human-like car-following model for autonomous vehicles considering the cut-in behavior of other vehicles in mixed traffic

Car-following is a common driving behavior which has a significant effect on driver safety and comfort. Although a large number of studies have focused on car-following models for autonomous vehicles (AV) and connected vehicles (CV), car-following models for AV and CV which consider cut-ins in mixed traffic have not been investigated. In this study, a human-like car-following model for AV and CV was developed by examining the effect of cut-in vehicles on car-following behavior and the expectations of drivers. The cut-in position, reaction time, acceleration, and desired distance were investigated on a real freeway in an instrumented vehicle. Corresponding to results from previous studies, the cut-in vehicles maintain a safe distance from the preceding vehicle and a larger distance from the following vehicle to avoid conflict. Analysis of the behavior of the following driver illustrates that in the keeping stages, the reaction time after the cut-in is 0.85 s for the acceleration stimulus and 0.70 s for the deceleration stimulus. These times are shorter than the response time before the cut-in for the acceleration (1.95 s) and deceleration stimuli (1.66 s). The acceleration, rate of increase in the acceleration with the relative speed, and the desired distance are lower after than before the cut-in events. In this paper, a human-like car-following model for cut-in situations is proposed, which is designed for autonomous vehicles. Unlike previous car-following models, the proposed model has a shorter response time and lower deceleration in cut-in situations. The proposed model may help to improve car-following safety, driver comfort, and trust in AVs and CVs.

A car-following model accounting for the driving habits

In this paper, we propose a car-following model to ensure the car-following safety, stability, comfort and fuel-optimal with consideration of the driver’s different driving habits under different headways. During the car-following state, the following vehicle’s acceleration is determined by its driving strategy in different headways to fit the driving habits. When the headway is overlong, the following vehicle will shorten the headway by a comfort and fuel-optimal acceleration–uniform motion–deceleration process to ensure the stable car-following behavior. When the headway is too short, the following vehicle take emergency brake to ensure the car-following safety. When the headway is within the desired range, the following vehicle can run at a relatively steady speed since the comfort driving is an important reason for drivers to adopt a new driving behavior. The numerical results show that the proposed model can ensure the safety, stability, comfort, fuel-optimal and the consistency with the driver’s driving habits under the car-following state.

Impact of heterogeneity of car-following behavior on rear-end crash risk

An increasing number of vehicles travel on freeways result not only in traffic congestions but also accidents. Rear-end crashes in freeways can be collectively attributed to drivers, vehicles, and road infrastructure, but driving behavior plays a key role in influencing car-following safety. This study aims to investigate the impact of heterogeneity of driving behavior on rear-end crash risk. Driving behavior depends on perceived risk levels, acceleration and deceleration habits, and driver reaction characteristics. Thus, the influencing factors of rear-end crash risk were initially analyzed by using the desired safety margin (DSM) model. Subsequently, five driving behavior parameters, including upper and lower limits of DSM, sensitivity coefficients of acceleration and deceleration, and response time, were calibrated by using the vehicle trajectories from the Next Generation Simulation I-80 datasets. Simulation experiments were designed to evaluate the impact of heterogeneity of car-following behavior on rear-end crash risk. Results showed that decreasing the lower (or upper) limit of the DSM, increasing the response time, increasing the sensitivity coefficient for acceleration, or decreasing the sensitivity coefficient for deceleration can increase rear-end crash risk. In addition, if stable and unstable driving styles coexist, then their proportions have important influences on rear-end crash risk. These results imply that two critical factors affect shock waves, namely, driving behavior characteristics and proportion of different driving styles. Thus, a potential strategy for the adjustment of the proportions of unstable driving styles can attenuate shock waves and reduce rear-end crash risk to a certain extent. Moreover, a wide extent of driving behavior heterogeneity can attenuate shock waves and subsequently reduce rear-end crash risk. Overall, driving behavior heterogeneity has an important impact on rear-end crash risk. Exploring the effect of each driving behavior parameter on rear-end crash probability is useful for urban road traffic control, and it can provide improved understanding of abnormal driving behavior characteristics to minimize rear-end crash risks.

Parsimonious shooting heuristic for trajectory design of connected automated traffic part II: Computational issues and optimization

Advanced connected and automated vehicle technologies enable us to modify driving behavior and control vehicle trajectories, which have been greatly constrained by human limits in existing manually-driven highway traffic. In order to maximize benefits from these technologies on highway traffic management, vehicle trajectories need to be not only controlled at the individual level but also coordinated collectively for a stream of traffic. As one of the pioneering attempts to highway traffic trajectory control, Part I of this study (Zhou et al., 2016) proposed a parsimonious shooting heuristic (SH) algorithm for constructing feasible trajectories for a stream of vehicles considering realistic constraints including vehicle kinematic limits, traffic arrival patterns, car-following safety, and signal operations. Based on the algorithmic and theoretical developments in the preceding paper, this paper proposes a holistic optimization framework for identifying a stream of vehicle trajectories that yield the optimum traffic performance measures on mobility, environment and safety. The computational complexity and mobility optimality of SH is theoretically analyzed, and verifies superior computational performance and high solution quality of SH. A numerical sub-gradient-based algorithm with SH as a subroutine (NG-SH) is proposed to simultaneously optimize travel time, a surrogate safety measure, and fuel consumption for a stream of vehicles on a signalized highway section. Numerical examples are conducted to illustrate computational and theoretical findings. They show that vehicle trajectories generated from NG-SH significantly outperform the benchmark case with all human drivers at all measures for all experimental scenarios. This study reveals a great potential of transformative trajectory optimization approaches in transportation engineering applications. It lays a solid foundation for developing holistic cooperative control strategies on a general transportation network with emerging technologies.

Realization of a Dilemma-Zone Guiding Algorithm at Signalized Intersections

This paper presents a dilemma-zone (DZ) avoidance-guiding system for vehicles approaching an intersection. The purpose of the system is to assist drivers in determining the driving behavior and to prevent vehicles from being caught in a DZ at the onset of a yellow phase. The optimal driving behavior is determined through warning information or a detailed guiding strategy. To calculate the guiding strategies, a DZ-guiding algorithm is proposed with a special focus on the vehicle DZ state and the interaction between vehicles. A simulation-based study proved the function of the proposed system and the effectiveness of the algorithm. It is found that, based on the conditions of driver's comfort and car-following safety, the guiding system can provide proper guidance for vehicles and can determine the optimal driving behavior in advance.