Autonomous vehicles (AVs) are being introduced to the traffic system with the promise of improving current traffic status. However, the empirical data also indicate contrary effects with estimated higher crash rate and change of crash patterns. Therefore, it is necessary to investigate the driving behavior of AVs and human-driven vehicles (HDVs) in real mixed traffic. Current studies have analyzed the driving behavior of AVs and HDVs, as well as behavioral adaptations of drivers of HDVs based on empirical data. While they play an important role in traffic systems, signalized intersections have not been studied sufficiently in this context. Therefore, this study aims to utilize the Waymo open dataset to characterize and quantify the behavioral differences of AVs and HDVs at signalized intersections. Five parameters of driving behavior related to signalized intersections were characterized according to five critical maneuver phases, which were identified by wavelet transform and threshold-based method. Statistically significant differences in driving behavior between AVs and HDVs were found, from three categorized situations: vehicle approaching the red light/queue, vehicle responding to the green light (as the first vehicle), and vehicle responding to its preceding vehicle (in the queue). Further, behavioral adaptations of HDV drivers were revealed in that they tended to keep closer to the stopped AVs in a queue and to react more strongly to AV start-up maneuvers when the traffic light turns to green.

@en

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.

@en

The urban expressway system serves as a key role in the roadway transportation system. It provides an efficient and comfortable approach for long-distance travel within the city. However, the safety status of the urban expressways is becoming a critical issue as the high-frequent traffic crashes have severely influenced the traffic operations. Among the safety influencing factors, including traffic operational parameters (such as traffic speed and volume), geometric features and traffic participants’ characteristics (such as vehicle roadway usage patterns), the traffic operational parameters and geometric features have been widely investigated. However, the impacts of traffic participants’ characteristics on traffic safety have never been examined. This unprecedented study aims to link vehicles’ roadway usage patterns with traffic safety through crash frequency analyses. First, the roadway usage patterns were identified using Latent Class Cluster Analysis (LCCA) based on their traveling rates. Then, the hourly-based crash frequency analysis data were formulated with traffic operational parameters, geometric features and crash data. Finally, crash frequency analysis models were developed to unveil the relationships between the crash occurrence and their influencing factors. The modeling results showed that the Random Effects Hurdle Negative Binomial Model (REHNBM) provided better goodness-of-fit. And it concluded that higher proportions of vehicles with low-level roadway usage pattern would substantially enhance the possibility of crash occurrence; while the proportions of vehicles with the medium-high-level roadway usage pattern had negative impacts on crash occurrence probability. Finally, safety improvement recommendations and strategies based on the modeling results were put forward.@en

The real-time crash risk analyses were proposed to establish the relationships between crash occurrence probability and pre-crash traffic operational conditions. Given its great application potentials that link with Active Traffic Management System (ATMS) for proactive safety management, it has become an important research area. Currently, researchers mainly developed the real-time crash risk analysis models with traffic flow descriptive statistics employed as explanatory variables and with re-sampled balanced dataset, which hold the limitations of insufficiently capturing the temporal-spatial traffic flow characteristics and failing to provide classification capabilities when deal with the imbalanced datasets. In this study, a Convolutional Neural Network (CNN) modelling approach with refined loss functions has been first time introduced to the real-time crash risk analyses. The primary objectives of the proposed CNN models are: (1) utilizing the tensor-based data structure to explore the multi-dimensional, temporal-spatial correlated pre-crash operational features; and (2) optimizing the loss functions to overcome the low classification accuracy issue brought by the imbalanced data. Data from the Shanghai urban expressway system were utilized for the empirical analysis. And a total of three types of loss functions, including traditional binary cross entropy, the α-weighted cross entropy and the focal loss, were introduced and being tested with varying ratios of crash and non-crash datasets. The modeling results show that the CNN model has better classification performance compared to the traditional Multi-layer Perceptrons (MLP) model with the tensor-based structure data. Besides, the developed CNN model with focal loss function has substantial classification enhancement under the imbalanced datasets. Finally, the distributions of predicting probabilities for balanced and imbalanced datasets were plotted to understand the effects of the imbalanced dataset and revealed how the proposed CNN model with focal loss function improves the model performance.@en

Crash frequency prediction models have been an important subject of safety research that unveils a relationship between crash occurrences and their influencing factors. Recently, the hourly-based refined-scale crash frequency analysis becomes attractive since it holds the benefits of introducing time-varying explanatory information (e.g. traffic volume and operating speed). However, crash frequency data with short time intervals possess the analytical issues of excessive zeros and unobserved heterogeneity. In this study, a marginalized random effects hurdle negative binomial (MREHNB) model was, for the first time, developed in which the hurdle modelling structure handles the excessive zeros issue and site-specific random effect terms capture the factors associated with unobserved heterogeneity. Moreover, the marginalized inference approach was introduced here to obtain the marginal mean inference for the overall population rather than subject-specific estimations. Empirical analyses were conducted based on data from the Shanghai urban expressway system, and the MREHNB model was compared with the HNB (hurdle negative binomial) and the REHNB (random effects hurdle negative binomial) model. In terms of model goodness-of-fits, REHNB and MREHNB model showed substantial improvement compared to the HNB model while there was no distinct difference between the REHNB and MREHNB models. However, as for the estimated parameters, the MREHNB model provided better inference precisions. Furthermore, the MREHNB model provided interesting findings for the crash contributing factors, for example, higher ratios of local vehicles within the traffic volume would enhance the probability of crash occurrence; and a non-linear relationship was concluded between traffic volume and crash frequency with the moderate level of volume held the highest crash occurrence probability. Finally, in-depth analyses about the modeling results and the model technique were discussed. The results will assist in designing more efficient control strategies for near real-time traffic management.@en