According to the results of a survey administered by the World Health Organization, the primary cause of the majority of traffic accidents is risky driving behavior. Consequently, the identification of risky driving behaviors is imperative for ensuring road safety. In order to effectively address the limitations of existing methods in modeling long-term dependencies, integrating external scenario information, and adapting to different driver personalities, this paper proposes a Transformer-based cross-attention feature fusion model for risky driving behavior recognition, referred to as CAFFT. This model utilizes a hierarchical Transformer encoder structure to capture fine-grained changes and long-term dependencies in driving behavior. Furthermore, the cross-attention mechanism, a cross-modal attention mechanism, is introduced to effectively fuse driving behavior features with weather-related external scenario features, thereby rendering the research scenario more aligned with real-world driving conditions. Finally, through cross-validation and cross-test set generalization experiments, the universality and stability of the method are verified. The experimental results on real driving datasets demonstrate that the CAFFT model achieves substantial performance enhancements in recognizing risky driving behavior. This improvement leads to enhanced accuracy and reliability in recognizing risky driving behavior, thereby providing substantial support for the development of intelligent driving assistance systems.

In the evolving landscape of mixed traffic environments, the interaction between Connected and Autonomous Vehicles (CAVs) and Human-driven Vehicles (HVs) introduces complex dynamics that challenge traditional traffic control paradigms. The Mixed Platoon Hierarchical Control (MPHC) model innovatively addresses these dynamics by integrating advanced control strategies that optimize the coexistence of CAVs and HVs. Specifically, the model introduces two novel approaches: the Incorporate Real-time Changes in Dynamic Headway (IRC_DH), which dynamically adjusts CAV headways by utilizing real-time traffic data and road conditions, enhancing platoon efficiency and reducing headway by up to 24.35%; and the Incorporate Controls to Improve Response Speed (IC_IRS) for HVs, a multi-variable control strategy that considers vehicle states, reducing speed differentials by as much as 53.125%, thereby stabilizing platoon dynamics and significantly lowering collision risks. These strategies distinguish themselves from existing approaches by providing a more precise, adaptable, and robust solution to the dynamic challenges of mixed traffic flow, addressing both efficiency and safety. Simulation results confirm that these strategies significantly improve traffic flow efficiency and safety, providing a scalable and adaptable framework for traffic management in mixed vehicular environments. This work makes a substantial contribution to the field by emphasizing a comprehensive, real-time, and multi-dimensional approach to managing mixed traffic environments, thereby advancing the state-of-the-art in CAV-HV integration.

Pedestrian detection plays a crucial role in autonomous driving systems. To ensure reliable and effective detection in challenging conditions, researchers have proposed RGB–T (RGB–thermal) detectors that integrate thermal images with color images for more complementary feature representations. However, existing methods face challenges in capturing the spatial and geometric correlations between different modalities, as well as in assuming perfect synchronization of the two modalities, which is unrealistic in real-world scenarios. In response to these challenges, we present a new deformable-attention-based approach for weakly aligned RGB–T pedestrian detection. The proposed method uses a dual-branch cross-attention mechanism to capture the inherent spatial and geometric correlations between color and thermal images. Furthermore, it incorporates positional information for each image pixel into the sampling offset generation to enhance robustness in scenarios where modalities are not precisely aligned or registered. To reduce computational complexity, we introduce a local attention mechanism that samples only a small set of keys and values within a limited region in the feature maps for each query. Extensive experiments and ablation studies conducted on multiple public datasets confirm the effectiveness of the proposed framework.

Pedestrian trajectory prediction, which aims at predicting the future positions of all pedestrians in a crowd scene given their past trajectories, is the cornerstone of autonomous driving and intelligent transportation systems. Accurate prediction and fast inference are both indispensable for real-world applications. In this paper, we propose a stochastic non-autoregressive Transformer-based multi-modal trajectory prediction model to address the two challenges. Specifically, a novel graph attention module dedicated to joint learning of social and temporal interaction is proposed to explore the complex interaction among pedestrians while integrating sparse attention mechanism, pedestrian identity, and temporal order contained in the trajectory data. By doing so, the interaction across temporal and social dimensions can be simultaneously processed to extract abundant context features for prediction. Besides, to accelerate inference speed, we put forward a stochastic non-autoregressive Transformer model with multi-modal prediction capability where each future trajectory can be inferred in a parallel fashion, therefore, resulting in diverse trajectory predictions and less computational cost. Extensive experiments and ablation studies are performed to evaluate our approach. The empirical results demonstrate that the proposed model not only produces high prediction accuracy but also infers with fast speed. The code of the proposed method will be publicly available at https://github.com/xbchen82/SNARTF.

This paper proposes a new avoidance control model for automatic vehicle in facing dangerous lane-changing behavior. Firstly, the new lane-changing probability factor based on Gaussian-mixture-based hidden Markov model is constructed to predict the lateral-vehicle lane-changing probability and output the pre-control parameters. Secondly, the back propagation neural network avoidance model, which combined with driver’s avoidance behavior, is developed for achieving the instantaneous collision avoidance control. Moreover, the optimal solution between control parameters and vehicle stability is obtained by using linear quadratic regulator. Finally, the accuracy of the avoidance model is verified by the semi-physical driver-in-the-loop simulation based on PreScan/Simulink. Results show that the automatic vehicle with the proposed avoidance model can accurately and effectively take pre-braking and micro-steering behavior. The proposed model can greatly reduce vehicle collision probability and effectively take both safety and comfort of collision avoidance into account. In addition, the robustness of the control model under different network penetration is discussed.

Cooperative adaptive cruise control (CACC) is a promising technology for vehicle platoons to increase roadway capacity. This paper proposes a networked Lagrange system (NLS) based CACC dynamic model based on which a hierarchical anti-disturbance controller is developed to solve the stability problem of CACC in vehicle platoons with time-varying parameter uncertainties, external disturbances, and directional dynamic communication topology. First, a hierarchical anti-disturbance controller for NLS is constructed which comprises three layers, these are, the adaptive smooth estimator-control layer, the distributed classifying amplitude-related layer, and the anti-disturbance control layer, in which the parameter uncertainties are estimated in the adaptive smooth estimator-control layer, the disturbance classification and distribute control are executed in the distributed classifying amplitude-related layer and the anti-disturbance control layer respectively. In addition, the proposed controller is extended to address the stability problem of CACC in vehicle platoons with time-varying parameter uncertainties, external disturbances, and directional dynamic communication topology conditions, which shows the versatility of the controller. Finally, comparison studies and simulation results are provided to demonstrate the effectiveness, significance, and advantages of the presented controllers.

Vehicular networks are becoming a prominent research field in the intelligent transportation system (ITS) due to the nature and characteristics of providing high-level road safety and optimized traffic management. Vehicles are equipped with the heavy communication equipment which requires a high power supply, on-board computing device, and data storage devices. Many wireless communication technologies are deployed to maintain and enhance the traffic management system. The ITS is capable of providing services to the traffic authorities and precautionary measures to the drivers and passengers. Several methods have been proposed for discussing the security and privacy issues for the vehicular ad hoc networks (VANETs) and vehicular cloud computing (VCC). They receive a great deal of attention from researchers around the world since they are new technologies, and they can improve road safety and enhance traffic flow by utilizing the vehicles resources and communication system. Firstly, the VANETs are presented, including the basic overview, characteristics, threats, and attacks. The location privacy methodologies are elaborated, which can protect the confidential information of the vehicle, such as the location detail and driver information. Secondly, the trust management models in the VANETs are comprehensively discussed, followed by the comparison of the cryptography and trust models in terms of different kinds of attacks. Then, the simulation tools and applications of the VANETs are discussed, and the evolution is presented from the VANETs to VCC in the vehicular network. Thirdly, the VCC is discussed from its architecture and the security and privacy issues. Finally, several research challenges on the VANETs and VCC are presented. In sum, this survey comprehensively covers the location privacy and trust management models of the VANETs and discusses the security and privacy issues in the VCC, which fills the gap of existing surveys. Also, it indicates the research challenges in the VANETs and VCC.

Recently, vehicular ad hoc networks (VANETs) embark a great deal of attention in the area of wireless and communication technology and are becoming one of the prominent research areas in the intelligent transportation system (ITS) because they provide safety and precautionary measures to the drivers and passengers, respectively. VANETs are quite different from the mobile ad hoc networks (MANETs) in terms of characteristics, challenges, system architecture, and their application. In this paper, we summarize the recent state-of-the-art methods of VANETs by discussing their architecture, security, and challenges. Secondly, we discuss the detailed analysis of security schemes and the possible measures to provide secure communication in VANETs. Then, we comprehensively cover the authentication schemes, which is able to protect the vehicular network from malicious nodes and fake messages. Thus, it provides security in VANETs. Thirdly, we cover the mobility and network simulators, as well as other simulation tools, followed by the performance of authentication schemes. Finally, we discuss the comfort and safety applications of VANETs. In sum, this paper comprehensively covers the entire VANET system and its applications by filling the gaps of existing surveys and incorporating the latest trends in VANETs.

This paper presents a framework of accident prediction with a new perspective. First, the new framework of Chain of Road Traffic Incident (CRTI) is proposed, in which the observed vehicle movement features are viewed as road traffic system's external “performance” that, in essence, reflect the internal “health states” (safety states) of the system at a specific time. A two-stage modeling procedure of CRTI is then proposed using scenario-based strategy: 1) a support vector machine is utilized to classify leaving lane scene versus remaining in lane scene and 2) Gaussian-mixture-based hidden Markov models are developed to recognize accident versus non-accident pattern CRTI given the classified scene. Moreover, the application procedure of the CRTI framework to online collision prediction is proposed. Finally, a simulation test of a typical vehicle collision scene based on PreScan platform is designed and carried out for model training and validation, and has shown promising results in accident prediction using the proposed framework. The CRTI framework could provide a new foundation for developing early warning/intervention strategies in driver assistance system under complex traffic environments.