Autonomous driving systems require extensive testing under a wide range of traffic conditions to ensure safety and reliability. Powered two-wheelers (PTWs), such as motorcycles and scooters, are particularly challenging due to their agile, nonlinear, and sometimes abrupt maneuvers. This paper proposes a conditional denoising diffusion probabilistic model to generate high-risk PTW pre-crash trajectories from limited in-depth crash data. By reversing a controlled noise process, the model synthesizes PTW trajectories that match the kinematic characteristics of real pre-crash cases while covering a wide range of risky interactions. Experiments on reconstructed crash trajectories from the China In-depth Mobility Safety Study Traffic Accident (CIMSS-TA) database show that the proposed method produces trajectories with higher physical plausibility, closer similarity to real-world crashes, and richer high-risk coverage than baseline models. When integrated into a simulation platform, the generated scenarios trigger more frequent and more diverse PTW conflict situations than the original crash set, indicating improved coverage of safety-critical conditions for autonomous vehicle testing.

The effectiveness of vehicle collision avoidance systems depends on the precision of collision risk prediction models. However, current models often neglect the driver's condition, resulting in their poor ability to predict near-crash events triggered by aggressive, fatigued, or distracted driving. Additionally, current models overlook the differences in modality, type, and variability of multi-source data, leading to insufficient feature extraction from input data, which in turn limits the model's prediction accuracy. To address these issues, we developed an end-to-end pre-trained deep framework (PM-Transformer) with a Transformer, consisting of multi-module recurrent convolutional neural networks. The framework includes four modules: (1) pre-trained time series module that extracts spatiotemporal information from traffic time series using one-dimensional convolutional neural network - long short-term memory; (2) pre-trained spectral module that learns visual temporal representations from traffic spectrograms using two-dimensional convolutional neural network - long short-term memory; (3) metadata module for vectorizing traffic metadata; (4) fusion module that semantically integrates features from the three modules using a Transformer. Results show that the proposed model can achieve the same prediction accuracy as other models using only 5 % of their training sample size. Compared to other traditional models, our model improves the accuracy of risk prediction by 6 %, 9 %, and 4 %, respectively, with small sample sizes (0.5 s, 1 s, and 2 s in advance), while also maintaining the best performance with larger sample sizes. Findings of this study hold significant potential for improving the effectiveness of vehicle collision avoidance systems.

The unique structure and lighting conditions of freeway tunnels significantly raise the crashe risk. As connected-automated vehicles (CAVs) increasingly coexist with human-driven vehicles in mixed traffic environments, the complexity of tunnel traffic risk increases. This study introduces a novel CAVs-based dynamic speed limit (CAVs-based DSL) control strategy specifically designed for freeway tunnels in mixed traffic to enhance safety and efficiency. This strategy utilizes CAVs as moving barriers to implement DSL control, thereby reducing the risk of crashes and improving the efficiency in tunnels. This study models the CAVs-based DSL control as a safe reinforcement learning (SRL) problem and develops the augmented Lagrange multiplier method combined with the deep deterministic policy gradient (ALM-DDPG) algorithm to solve it. A comprehensive simulation environment based on real-world tunnel scenarios is developed to evaluate the effectiveness of the proposed strategy. The results show that the ALM-DDPG-based DSL reduces traffic conflicts by 22.96%–35.68% and travel time by 15.34%–21.21%, compared to fixed speed limit control across different market penetration scenarios of CAVs, leading to significant safety and efficiency improvements. Compared with the conventional DDPG and the sum-weighted multiobjective DDPG algorithms, the ALM-DDPG algorithm yields more balanced performance in terms of traffic safety and efficiency. The ALM-DDPG algorithm also offers significant advantages over traditional Lagrange multiplier-based SRL algorithms, providing faster convergence and higher stability, especially in scenarios with high-CAV penetrations. This study highlights the significant potential of integrating SRL with CAV technology to address the complex safety challenges of freeway tunnels.

Compared to autonomous driving, cooperative driving offers higher efficiency and safety. However, the interaction based cooperative driving makes its control system more complex. To ensure the functional safety of this technology, extensive scenarios testing is required. Due to the complexity of cooperative driving scenarios and the scarcity of cooperative driving data, testing and validating it becomes challenging. Additionally, existing data generation methods are limited in their ability to generate effective cooperative driving scenarios. To address these challenges, this study develops a novel Diffusion-Transformer-Wavelet (Diffwavelet) model for cooperative driving scenario generations. A rule-based method is firstly proposed to extract cooperative scenarios, which integrates a surrogate safe measure of Stopping Distance Index (SDI) with a Dynamic Proactive Behavior Identification method (DPBI). Furthermore, the Diffwavelet model is developed for cooperative scenario generations, overcoming the limitations of existing generative models in generating realistic cooperative driving scenarios. The Wavelet module is employed to capture both local and trend features of the data, enhancing the models ability to capture the characteristics of cooperative driving data. Results show that, through the proposed rule-based method, a large number of cooperative driving scenarios can be extracted effectively. The proposed Diffwavelet model outperforms existing techniques for scenario generations, such as TimeGAN and TimeDiffusion, in terms of data quality, particularly in preserving the interaction characteristics of different vehicle trajectories. The parameter distribution of the generated scenario data closely matches that of the original data and also enhances scenarios diversity.

Deep reinforcement learning (DRL) has become a proactive control strategy at intersections in mixed-autonomy traffic of connected autonomous vehicles (CAV) and humandriven vehicles (HDV). Despite the superior performance of DRL in intersection application and vehicle efficiency optimization compared to traditional traffic signal control (TSC) methods, scant research focused on pedestrians, one key and vulnerable component of traffic participants, especially pedestrian safety. Based on DRL techniques and protected/prohibited right-turn (PPRT) policy that regulates vehicular right-turn movement for pedestrian protection, this study proposes a two-stage pedestrianvehicle TSC system called TS-PVL, which cooperative control the traffic signal and pedestrian-vehicle flows, aiming to minimize pedestrian-vehicle conflicts and delays. The first stage adopts the DRL algorithm frameset, comprising the state, action, and reward designed for pedestrians and vehicles, to schedule the green time of PPRT signals. The second stage implements a speed planning model based on the DRL-solved signal timing to manage pedestrian and vehicle movement. To evaluate TS-PVL effectiveness, a real-world intersection with daylong traffic flow data in Changsha City is simulated as the case study in mixedautonomy traffic using various CAV and HDV car-following models. Experimental results show that the proposed TS-PVL method reduces total delay by 15.6%, stop frequency by 17.2%, pedestrian-vehicle conflicts by 28.3%, and jaywalking incidents by 31.5% compared to current TSC and DRL-based TSC methods. Besides, as the market penetration rates of CAVs in mixedautonomy traffic increase, the efficacy of TS-PVL reaches its optimal level.

Scenario-based testing offers a robust method for evaluating the safety of Autonomous Vehicles (AVs). In this study, we introduce LSTM-CGAN to generate scenarios for assessing AV safety performance. First, we extract static, environmental, and dynamic elements from the China In-depth Mobility Safety Study-Traffic Accident (CIMSS-TA) dataset and reconstruct 630 real-world crashes to obtain pre-crash trajectories of the ego and objective vehicles. The proposed model then synthesizes samples by utilizing high-risk trajectories extracted from pre-crash trajectories, resulting in the generation of 6,300 trajectories, ten times the original dataset. The synthesized samples closely resemble real-world trajectories while enhancing diversity, with 61.9% of road segment and 55.5% of intersection trajectories scenes being classified as challenging. By integrating static and dynamic elements with the synthesized trajectories, critical scenarios for counterfactual simulations are generated. Finally, the Baidu Apollo, a well-known Autonomous Driving System (ADS), is tested in simulations using both generated and original scenarios. It is evidenced that the proposed method is effective in generating more safety-critical scenarios and that static and environmental factors have different impacts on AV performance. This study provides valuable insights into the unknown safety-critical scenario construction of AV safety testing, thereby facilitating the commercialization of AVs.

Safety is a paramount concern in the development of Autonomous Vehicles (AVs) within the automotive industry. Ensuring AVs’ reliable performance requires the development of varied and realistic testing scenarios. Identifying unknown hazardous scenarios presents a notable challenge in the testing process. To address a wider spectrum of unknown hazardous scenarios, this study introduces a novel methodology for generating safety-critical scenarios for AV safety testing based on in-depth crash data. First, we used the China In-depth Mobility Safety Study-Traffic Accident (CIMSS-TA) database to decompose crash scenarios into static and dynamic variable tuples. Subsequently, the time series Generative Adversarial Networks (TimeGAN) was used to capture temporal features and to generate safety-critical scenarios using the original dataset. The synthesized scenarios were then assessed in terms of similarity, usability, and risk. To further validate the proposed method, we compared it with other generative models, highlighting the superior performance of TimeGAN. Finally, a well-known autonomous driving system, Baidu Apollo, was tested in original, synthesized, and naturalistic driving scenarios. Experimental results demonstrate that safety-critical scenarios generated from in-depth crash data preserve essential risk characteristics while increasing diversity. Baidu Apollo demonstrated superior safety performance in hazardous scenarios compared with human drivers, while naturalistic driving scenarios were less efficient in evaluating the performance of autonomous driving systems in safety-critical events. The synthesized scenarios provide a comprehensive dataset for testing AVs under high-risk conditions, overcoming the limitations of crash data and expanding the scope of safety-critical scenarios to improve AV technology evaluation and innovation.

Traffic accidents are a significant factor leading to injuries and property losses, prompting extensive research in the field of traffic safety. However, previous studies, whether focused on static environment assessment, dynamic driving analysis, pre-accident prediction, or post-accident rule checks, have often been conducted independently. Our introduces V2X Environmental Perception Multi-modal Large Model AccidentGPT for accident analysis and prevention. AccidentGPT establishes a multi-modal information interaction framework based on multisensory perception. It adopts a holistic approach to address traffic safety issues, providing environmental perception for autonomous vehicles to avoid collisions and maintain control. In human-driven vehicles, it offers proactive safety warnings, blind spot alerts, and driving suggestions through human-machine dialogue. Additionally, it aids traffic police and management agencies in considering factors such as pedestrians, vehicles, roads, and the environment for intelligent real-time analysis of traffic safety. The system also conducts a thorough analysis of accident causes and post-accident liabilities, making it the first large-scale model to integrate comprehensive scene understanding into traffic safety research. Project page: https://accidentgpt.github.io

The dedicated lanes management policy is a possible solution to the issues arising from the coexistence of human-driving vehicles (HDVs) and connected and autonomous vehicles (CAVs) in traffic. Although numerous studies have been conducted on the network deployment problem of CAV-dedicated lanes, the safety implications of CAV and its dedicated lanes are ignored. This study proposes a mathematical approach to optimize the deployment of CAV-dedicated lanes incorporating efficiency and safety concerns. An integrated framework is developed based on headway distributions to systematically evaluate the efficiency and safety performance of the road network. The platoon intensity index is utilized to model the platooning effect of CAVs on traffic safety and efficiency. A safety performance estimation method is proposed to account for the potential collision risk of mixed traffic flow and heterogeneity in car-following behavior. A bi-level programming model is adopted to solve the optimal deployment problem. The upper-level model is formulated as a bi-objective model to minimize the total travel time and the safety risk. The lower-level model describes the multi-class user equilibrium state of the CAV-HDV mixed traffic flow on the network. A genetic algorithm is utilized to solve the bi-level programming model and obtain the Pareto-optimal solution set. Two numerical studies are conducted to validate the proposed model and algorithm. The results revealed that the optimal deployment plan can significantly improve the road network’s safety and efficiency performance, whereas higher platoon intensities have a negative impact on traffic safety and efficiency for certain headway settings. Moreover, the results highlighted a trade-off between efficiency and safety in the optimal deployment problem, which may help decision-makers choose the optimal deployment plan based on the road network design needs.

Collision warning systems (CWSs) have been recognized as effective tools in preventing vehicle collisions. Existing systems mainly provide safety warnings based on single-directional approaches, such as rear-end, lateral, and forward collision warnings. Such systems cannot provide omni-directorial enhancements on driver’s perception. Meanwhile, due to the unclear and overlapped activation areas of above single-directional CWSs, multiple kinds of warnings may be triggered mistakenly for a collision. The multi-triggering may confuse drivers about the position of dangerous targets. To this end, this paper develops a spatial-state-based omni-directional collision warning system (S-OCWS), aiming to help drivers identify the specific danger by providing the unique warning. First, the operational domains of rear-end, lateral, and forward collisions are theoretically distinguished. This distinction is attained by a geometric approach with a rigorous mathematical derivation, based on the spatial states and the relative motion states of itself and the target vehicle in real time. Then, a theoretical omni-directional collision warning model is established using time-to-collision (TTC) to clarify activation conditions for different collision warnings. Finally, the effectiveness of the S-OCWS is validated in field tests. Results indicate that the S-OCWS can help drivers quickly and properly respond to the warnings without compromising their control over lateral offsets. In particular, the probability of drivers giving proper responses to FCW doubles when the S-OCWS is on, compared to when the system is off. In addition, the S-OCWS shortens the responses time of nonprofessional drivers, and therefore enhances their safety in driving.

To ensure the safety of Autonomous Vehicles (AVs), thorough testing across virtual simulation environments, closed facilities, and public roads is essential. Scenario-based testing stands out as a crucial method for evaluating AVs, with a key focus on constructing appropriate testing scenarios. Given the vulnerability of Powered Two-Wheelers (PTWs) riders, it is essential to investigate typical and representative car-to-PTWs crash scenarios and validate AV system safety performance in such situations. This study introduces a new method for generating high-risk scenarios by extracting typical testing scenarios from real-world crashes, thereby enhancing the realism of testing conditions. To evaluate the safety performance of AV systems, a crash-based testing approach is proposed. First, 222 car-to-PTWs crashes were extracted from the China In-depth Mobility Safety Study-Traffic Accident (CIMSS-TA) database and the pre-crash trajectories of crash-involved parties were accurately obtained by reconstructing the crashes case-by-case. Second, utilizing the$k$-medoids algorithm based on the Hausdorff distance to cluster these trajectories, six typical pre-crash trajectory clusters were extracted. Third, six sets of high-risk scenarios were generated using parameter discretization and combination testing based on the cluster centers. Finally, we conducted safety testing on the black-box automated driving system, Baidu Apollo, using the SVL Simulator virtual simulation platform. We evaluated its performance by subjecting it to the six sets of high-risk scenarios generated in this study. The experimental results demonstrate that Apollo can operate safely in most high-risk scenarios, indicating that automated driving systems can handle crashes that some human drivers cannot avoid, thereby improve traffic safety.

This paper introduces a model to analyze route choice behavior of taxi drivers for finding next passenger in urban road network. Considering the situation of path overlapping between selected routes in the process of customer-searching, a mixed path size logit (MPSL) model is proposed to analyze route choice behaviors through considering spatio-temporal features of route including customer generation rate, path travel time, cumulative intersection delay, path distance, and path size. Specially, customer generation rate is defined as attraction strength based on historical pick-up records in the route, the intersection travel delay and path travel time are estimated based on large scaled taxi global positioning system (GPS) trajectories. In the experiment, the GPS data were collected from about 36000 taxi vehicles in Beijing at 30-s interval during six months. In the model application, an area of approximately 10 square kilometers in the center of Beijing is selected to demonstrate the effectiveness of the proposed model. The results indicated that the MPSL model could effectively analyze the route choice behavior in customer-searching process and express higher accuracy than traditional multinomial logit model and basic PSL model.