Object detection is a cornerstone of environmental perception in advanced driver assistance systems(ADAS). However, most existing methods rely on RGB cameras, which suffer from significant performance degradation under low-light conditions due to poor image quality. To address this challenge, we proposes WTEFNet, a real-time object detection framework specifically designed for low-light scenarios, with strong adaptability to mainstream detectors. WTEFNet comprises three core modules: a Low-Light Enhancement (LLE) module, a Wavelet-based Feature Extraction (WFE) module, and an Adaptive Fusion Detection (AFFD) module. The LLE enhances dark regions while suppressing overexposed areas; the WFE applies multi-level discrete wavelet transforms to isolate high- and low-frequency components, enabling effective denoising and structural feature retention; the AFFD fuses semantic and illumination features for robust detection. To support training and evaluation, we introduce GSN, a manually annotated dataset covering both clear and rainy night-time scenes. Extensive experiments on BDD100K, SHIFT, nuScenes, and GSN demonstrate that WTEFNet achieves state-of-the-art accuracy under low-light conditions. Furthermore, deployment on a embedded platform (NVIDIA Jetson AGX Orin) confirms the framework's suitability for real-time ADAS applications.

Reconstructing complete traffic flow time-space diagrams from vehicle trajectories offer a comprehensive view on traffic dynamics at arterial intersections. However, obtaining full trajectories across networks is costly, and accurately inferring lane-changing (LC) and car-following behaviors in multi-lane environments remains challenging. This study proposes a generative framework for arterial vehicle trajectory reconstruction that jointly models lane-changing and car-following behaviors through physics-informed multi-task joint learning. The framework consists of a Lane-Change Generative Adversarial Network (LC-GAN) and a Trajectory-GAN. The LC-GAN models stochastic LC behavior from historical trajectories while considering physical conditions of arterial intersections, such as signal control, geometric configuration, and interactions with surrounding vehicles. The Trajectory-GAN then incorporates LC information from the LC-GAN with initial trajectories generated from physics-based car-following models, refining them in a data-driven manner to adapt to dynamic traffic conditions. The proposed framework is designed to reconstruct complete trajectories from only a small subset of connected vehicle (CV) trajectories; for example, even a single observed trajectory per lane, by incorporating partial trajectory information into the generative process. A multi-task joint learning facilitates synergistic interaction between the LC-GAN and Trajectory-GAN, allowing each component to serves as both auxiliary supervision and a physical condition for the other. Validation using two real-world trajectory datasets demonstrates that the framework outperforms conventional benchmark models in reconstructing complete time-space diagrams for multi-lane arterial intersections. This research advances the integration of trajectory-based sensing from CVs with physics-informed deep learning.

Traditional traffic control methods at signalised intersections primarily focused on optimising traffic signals without adequately addressing the coordination between signals and vehicles. Fortunately, with the advent of advanced vehicular communication technologies, real-time bidirectional communication between roadside infrastructure and vehicles has become feasible, significantly improving coordination between these elements. This paper presents a bi-level optimisation method for signalised intersections, enhancing arrive-on-green (AOG) performance in connected vehicles. By extending optimal control from a single dimension - whether spatial or temporal - to a two-dimensional spatial-temporal approach, we develop a comprehensive bi-level control framework. The framework includes outer-layer signal optimisation for maximising green utilisation and inner-layer platoon trajectory optimisation. Intermediate parameters and extended planning-time are proposed to improve solution finding. The effectiveness of the proposed joint optimisation method was evaluated through simulation case studies conducted in SUMO. The results showed increased efficiency and reduced stops, with stable, accurate control.

This paper presents an innovative joint traffic signal and trajectory control (JSTC) model for a mixed traffic environment with a low penetration rate of connected autonomous vehicles (CAVs). The JSTC model integrates a back-pressure (BP) signal controller that comprehensively considers the right-of-way with traffic modes and the person-based traffic demand, which aims to enhance intersection throughput and improve the probability of arrival on green (PAOG) of CAVs. Additionally, this paper provides a theoretical stability proof of the control system by deviating the upper bound of the Lyapunov drift penalty function. Furthermore, it simultaneously introduces a discretized CAV trajectory control model by discretizing continuous time variables with vehicle-to-infrastructure (V2I) communication periods for smooth and safe travelling. Simulation results demonstrate that the JSTC model exhibits excellent responsiveness and achieves the lowest convergence value of cycle maximum queue length across various flow-based traffic demand levels. The smoothness of the Lyapunov function values validates the ability to stabilize the control system. Two months of connected transit buses (CTB) operation data shows that substantial progress has achieved. These include the elimination of unnecessary stops, an over 20km/h increase in daily average speed, a 17% reduction in weekly average delay, and at least 10% reduction in weekday travel time.

Automated vehicles (AVs) or driverless taxis have already been operated in urban environments. However, understanding the social interactions between AVs and pedestrians at uncontrolled crosswalks remains limited. Specifically, how could human drivers balance driving safety and efficiency while effectively communicating their intentions to pedestrians? To address this research gap, this study conducted an in-depth analysis of humandriven vehicle (HDV)-pedestrian interactions at actual uncontrolled mid-block crosswalks by coupling HDV kinematics with pedestrian crossing behavior using the proposed time-to-conflict point and time-to-arrive (TTCPTTA) and space-time (S-T) diagrams. Drawing upon social value orientation assumptions as an analytical framework, we categorized HDVs' driving behaviors into five modes that balance safety and efficiency: egoistic, competitive (successful), competitive (unsuccessful), prosocial, and altruistic. The results revealed that in real scenarios, most interaction events fall into egoistic and competitive (successful) driving modes, with the altruistic driving mode being the least common. Moreover, the analyses showed that egoistic driving modes prioritize efficiency, while altruistic driving modes emphasize safety. Competitive (successful), competitive (unsuccessful), and prosocial driving modes offer different balances between safety and efficiency. Interestingly, it was found that HDVs may convey diverse social intentions to pedestrians through similar driving behavior among egoistic, competitive (successful), and competitive (unsuccessful) modes. Finally, this study discussed the practical implications of the proposed methods and results for AV driving behavior design and interaction safety.

Electric Vehicles (EVs) are cost-effective and widely recognized for their significant role in reducing greenhouse gas (GHG) emissions. However, concerns surrounding range anxiety and charge anxiety have hindered their widespread adoption. To address these concerns, traffic engineers have been working on developing control strategies to reduce energy consumption (EC). Unlike traditional gasoline-powered vehicles, EVs experience a notable increase in EC at speeds exceeding 25km/h. Consequently, minimizing EC often results in reduced speed and longer total travel time (TTT). In light of this, our paper proposes a novel trajectory and signal control method that leverages connected and automated vehicle (CAV) technology to realize a tradeoff between EC and TTT. Initially, the approach assumes all vehicles are connected and automated electric vehicles (CAEVs) capable of communication and coordination, to which a cooperative adaptive cruise control (CACC) model was applied. Then, the vehicles were controlled to avoid stops and achieve smoother trajectories at the intersections. Finally, the signal control was integrated to further reduce EC and TTT. The proposed method was evaluated with a simulation conducted in SUMO based on a busy corridor in the City of Edmonton, Canada. The developed method successfully balanced energy and traffic efficiency, reducing both EC and TTT by 14% and 38% respectively. More importantly, the computational burden of our method is considerably lighter compared to existing studies, making it highly suitable for real-time applications. Overall, the results presented in our study showcase the potential of achieving a more efficient and sustainable traffic system with the future existence of CAEVs.

This paper combines millimeter-wave radar (MMW) data with connected autonomous vehicle (CAV) trajectory data to estimate queue length on a second-by-second basis. Firstly, queued vehicles on multiple lanes with the same traffic movement are mapped to a virtual lane. Then, the presence of CAV or human-driven vehicle (HDV) for any given queueing index is determined. A Bayesian joint probability model is subsequently established for queue length expectation, considering the existence of the queued vehicle type as a condition. The average time headway and dissipation speed distribution are derived from departure timestamps, which allows for the calculation of the prior probability ratio of queue length. Lastly, the maximum likelihood estimation (EM) algorithm is employed for iteratively estimating the CAV penetration rate. Simulation results demonstrate the method offers a compelling trade-off between precision and second-based real-time performance, while field test results further confirm the wide applicability under various traffic conditions.

This paper conducts a comparative analysis of bidirectional communication topologies in vehicular platooning, emphasizing their impact on safety during travel. Introducing two novel metrics, the Accumulative Average Penalty of Minimum Time to Collision (AAPMTTC) and Accumulative Average Deceleration Rate to Avoid Collision (AADRAC), the study evaluates collision susceptibility between neighboring vehicles, taking into consideration their relative velocities and accelerations. Utilizing platoon dynamics based on intervehicle distances and their derivatives, the analysis captures the evolving vehicle behaviors over travel time. Results highlight the safety benefits of communication topologies where follower vehicles receive information from more preceding vehicles, particularly when incorporating the leader vehicle's state. This research underscores the significance of information exchange within vehicular platoons and offers insights for providing more safety through design of communication structure in automated driving scenarios in vehicular platooning.

With the capability of communicating with surrounding vehicles and infrastructures, connected and automated vehicles (CAVs) can safely drive closer with reduced headway, thereby potentially improving traffic efficiency. However, their superiority is compromised in the mixed traffic environment because of the interruption of human-driving vehicles (HDVs). In this circumstance, researchers proposed to physically separate CAVs and HDVs by deploying CAV-dedicated lanes (CAV-DLs). Nevertheless, the CAV-DLs may be underutilized, especially in low CAV penetration rate (PR) cases which may even reduce traffic efficiency. To solve this problem, two novel strategies were proposed in our study to better manage the CAV-DLs and magnify the benefit of CAVs: The first one is to dynamically allocate the right-of-way for CAV-DLs based on the predicted CAV-DLs’ effective utilization rate so that the HDVs can be allowed to use the dedicated lanes when they are not adequately occupied. The second strategy is motivated by the economic instrument, which allows HDVs to use the CAV-DLs by paying a toll. The toll is determined by the travel time difference between CAV-DL and general lane (GL), and these tolls can be utilized as subsidies to stimulate drivers to purchase CAVs for promoting their adoption. The two strategies were evaluated using the case study designed based on the network of Edmonton downtown area in Canada, and the results demonstrated that both methods can significantly reduce travel time. Besides, the two strategies were compared comprehensively in terms of their effectiveness and policy enforceability, which can provide some guidance for both traffic policymakers and practitioners.

Vehicular platooning, a configuration comprising a leading vehicle and multiple follower vehicles (FVs) seeks to achieve and maintain specific intervehicle distances (IDs) while synchronizing FVs with the velocity and acceleration of the leading vehicle. Before attaining a desired stable state, the IDs may undergo transient fluctuations. While the attainment of internal stability is pivotal for realizing the intended spacing between vehicles, it does not inherently guarantee that these transient fluctuations remain within safe thresholds, thereby mitigating the risk of collisions. Communication between vehicles has a critical role in vehicular platooning and significantly influences these transient distance fluctuations. Consequently, we present a mapping between the initial conditions and these transient fluctuations which hinges on the communication topology (CT), as well as the control parameters. Specifically, our focus is directed toward bidirectional CTs (BDCTs), wherein FVs possess the capability to communicate both with preceding and subsequent vehicles within the platoon. Investigation of these mappings illuminates the advantages and disadvantages of various BDCTs. Notably, we discern that within BDCTs, the receipt of information from a greater number of vehicles situated behind may at times hinder the overall performance of the platoon, resulting in larger deviations from the desired IDs or the velocity and acceleration of the leading vehicle. In contrast, information derived from vehicles located ahead, particularly the leading vehicle itself, serves to enhance IDs and therefore contributes significantly to the safety of the platoon. In conclusion, our theoretical insights are substantiated through a series of simulations.

In recent years, vehicle trajectory data has become increasingly available from connected vehicles (CVs). CVs, acting as mobile sensors, can cover almost every intersection and provide enriched traffic information. In this context, this study proposed a novel statistical model-based method to predict arterial travel time using CV trajectories. The queue clearing time during green time is assumed to follow a Gamma distribution, and a maximized log-likelihood estimation (MLLE) is utilized to calculate related parameters. A hierarchical framework is further developed to improve both prediction accuracy and efficiency. First, the cloud (i.e., Traffic Management Center [TMC]) estimates the CV penetration rate (PR) and calibrates necessary model parameters offline. Then, the network edge (i.e., Mobile Edge Computing [MEC]), conducts the prediction online. A route in the City of Edmonton, Canada, is simulated to test the proposed method. The simulated CVs' trajectories are collected to estimate the PR and the cycle-by-cycle queue length at intersections. After that, the MEC at each intersection conducts travel time prediction with the parameters obtained from the TMC. The results achieved a low Root Mean Square Error (RMSE) of travel time prediction, averaging 0.9 minutes. Besides, the running time for a one-hour online prediction only costs 2.3 seconds.

Signal coordination is vital for smooth vehicle flow in corridors. Evaluating arterial progression typically relies on performance indices like travel time and delay, as well as tools such as the Purdue Coordination Diagram (PCD). However, these methods often overlook trajectory details and fail to capture continuous travel patterns. To quantify and visualize the arterial progression quality, this paper identifies spatially distinct vehicle groups using trajectory data and investigates the temporal travel patterns of each group. A velocity matrix of vehicles' travel along a corridor is extracted from the trajectories. Then a nonnegative matrix factorization (NMF) is utilized to reduce the dimensionality of the velocity matrix, which is decomposed into two matrices representing vehicle travel patterns and their weights. The resulting eigentravel pattern matrix indicates traffic flow smoothness and identifies co-occurrences of arrivals on green (AOG) or red (AOR) signals. Additionally, a novel metric called the Proportion of Arrival on Green (P-AOG) evaluates progression quality under different traffic control systems. The trajectory data are generated by the simulation software SUMO, and the simulation is based on six consecutive intersections of 23rd Avenue in Edmonton, Canada. A case study involving a before-and-after comparison of two signal control strategies is used for demonstration.

Existing signal control systems for urban traffic are usually based on traffic flow data from fixed location detectors. Because of rapid advances in emerging vehicular communication, connected vehicle (CV)-based signal control demonstrates significant improvements over existing conventional signal control systems. Though various CV-based signal control systems have been investigated in the past decades, these approaches still have many issues and drawbacks to overcome. We summarize typical components and structures of these existing CV-based urban traffic signal control systems and digest several important issues from the summarized vital concepts. Last, future research directions are discussed with some suggestions. We hope this survey can facilitate the connected and automated vehicle and transportation research community to efficiently approach next-generation urban traffic signal control methods and systems.

The rapid development of Internet of Vehicles (IoV) data powers various online intelligent transportation applications, such as network travel time reporting. However, the accuracy might be severely compromised due to limited probe vehicle sampling frequency. On that account, this article proposes a dynamic multigraph model-enabled framework to estimate reliable network travel time, even in low-IoV-frequency arterial corridors. The proposed framework first develops an improved sparse IoV travel time decomposition method. The segment travel time is further divided into the free-flow running time and static and dynamic delays. Second, a dynamic multigraph traffic network model (DMGTN) is developed to aid the proposed decomposition method. The model analyzes complicated spatiotemporal relevance between segments from multiple perspectives: the real-time travel time, congestion level, signal control (which is frequently neglected in previous research), and segment properties. Additionally, two distinct enhanced modules are designed for handing dense and sparse network graphs, respectively. This allows for a more efficient inspection over large-scale intricate arterial networks while maintaining precision. Field implementation is conducted in the downtown area of Zhangzhou, China. Compared to other high-performance baseline models, the designed DMGTN model as well as the proposed decomposition method demonstrate state-of-the-art accuracy and successfully capture travel time variability. The proposed framework better utilizes available IoV data to provide valuable traffic information for commuters and traffic management agencies.

Green Light Optimal Speed Advisory (GLOSA) is an intelligent transportation system technology that uses real-time information from traffic lights and real-time vehicle location and speed information to advise drivers on the optimal speed, so that it can reach the optimal speed before the next traffic light turns green. It aims to reduce vehicle energy consumption and traffic congestion while improving road safety. In this paper, a two-layer frame is proposed to optimize vehicle speed as to reduce vehicle energy consumption while vehicles pass through intersections without stopping. The upper frame is to calculate the passable time area of vehicles arriving at the intersection without stopping by predicting the queuing information at intersections. The lower frame is to calculate the energy consumption at each point in the passable time area when vehicles arrive at the intersection, and the Dijkstra algorithm is used to solve the path where vehicles pass through continuous intersections without stopping and have the least energy consumption. The simulation results show that, compared with the constant speed strategy through the intersection, the proposed multi-objective optimization strategy can reduce energy consumption by 0.7%, 2.6% and 9.8% under the condition of over-saturated, saturated and under-saturated traffic volumes, respectively. It is an effective vehicle speed guidance optimization strategy.

Work zones, being a critical component of roadway transportation systems, can benefit greatly from computer vision-enabled roadway infrastructures, specifically in connected vehicle (CV) environments. Connected infrastructures, such as roadside units (RSU) and on-board units (OBU), can greatly improve the environmental awareness and safety of CVs driving through a work zone. The contribution of this paper lies in developing a vision-based approach to generate work zone safety messages in real time, utilizing video streams from roadside monocular traffic cameras that can be used by CV work zone safety apps on mobile devices to reliably navigate through a work zone. A monocular traffic camera calibration method is proposed to establish the accurate mapping between the image plane and Global Position System (GPS) space. Real test scenarios show that our algorithm can precisely and effectively locate work zone boundaries from a monocular traffic camera in real time. We demonstrate the capabilities and features of our system through real-world experiments where the driver cannot see the work zone. End-to-end latency analysis reveals that the vision-based work zone safety warning system satisfies the active safety latency requirements. This vision-based work zone safety alert system ensures the safety of both the worker and the driver in a CV environment.

Simultaneous localization and mapping (SLAM) has been widely used in augmented reality (AR), virtual reality (VR), robotics, and autonomous vehicles as the theoretical basis for robots to perceive their environment. Most popular SLAM algorithms assume that objects in the scene are static. Solving dynamic problems in SLAM is now attracting increasing attention. In this article, we propose a method that combines semantic segmentation information and spatial motion information of associated pixels to cope with dynamic objects based on ORB-SLAM2. We add a deep segmentation network SegNet to segment input image and obtain the semantic information for each feature point. Next, the spatial velocity of feature points between adjacent frames is calculated assuming uniform motion. Finally, the two parts are fused for the final judgment, and the dynamic feature points are removed to improve positioning accuracy. We evaluate our SLAM algorithms using the public KITTI dataset. The proposed algorithm has a similar overall accuracy level to ORB-SLAM2, but it is more accurate in sequences with many dynamic objects. On KITTI’s raw data sequence containing multiple dynamic objects, our pipeline achieves the best performance, improving 39.5% compared with the original ORB-SLAM2 system. We compare our algorithm with other state-of-the-art SLAM systems used to cope with dynamic environments. The results show that the proposed algorithm has a better performance.

Forecasting the forthcoming intersection movement-based traffic volume enables adaptive traffic control systems to dynamically respond to the fluctuation of traffic demands. In this paper, a deep-learning based Signal-control Refined Dynamic Traffic Graph (ScR-DTG) Model is proposed for advancing the network-level movement-based traffic volume prediction task. The proposed model attempts to further improve the-state-of-art and practice algorithms in traffic prediction for arterial network adaptive signal control utilizing tradition traffic flow theory boosted deep-learning methodology. For precisely inferencing the movement-based demand at cycle-to-cycle level, the proposed model incorporates spatial graph convolution inferencing layer and temporal inferencing layer to explore both the intricate spatial temporal dependencies, respectively. A signal control refining module is contrived to deduce the controlled movement saturation flow and introduce the essential control inferences, which is of great significance but frequently neglected in the previous researches. Additionally, according to the real-time movement specified travel time, this paper creatively constructs an adjacent graph with dynamic order for more accurately capturing the ever-changing spatial relevancies. Field experiments with multiple signal schemes were conducted in the downtown area of Zhangzhou (China). The promising results demonstrated the state-of-the-art accuracy than other high-performance volume prediction algorithms. Implementing the proposed model enables to obtain accurate movement-based volume predictions, which would assist the traffic management agencies in adjusting signal timing adaptively and further improve the efficiency of signal intersection.

We develop and assess centralized and decentralized signal control systems with short-term origin-destination (OD) demands as inputs. Considering each intersection turning movement as a virtual link, we assign traffic demand to paths based on minimal instantaneous travel time. Then, the optimal control is formulated using a G/G/n/FIFO open queueing network model (QNM). We also solve the issue of optimal control using a three-step naïve method for the centralized system with the new inputs. Because the optimization of large-scale network traffic signals can involve sizeable numbers of decision variables and nonlinear constraints, making it a nondeterministic polynomial time (NP) complete problem, we further decompose the centralized system into a decentralized system where the network is divided into subnetworks. Each subnetwork has a dedicated agent that optimizes signals within it. Furthermore, traffic demand for the entire network is decomposed into demands for subnetworks via path decomposition index (PDI). The proposed control systems are applied to test scenarios constructed using different demand profiles in grid networks. We also investigate the impact of network decomposition strategy on signal control system performance. Results show that network decomposition with smaller subnetworks results in less computational time (CT) but increased average travel time (ATT) and total travel delay (TTD).

Pushbutton control is ideal for midblock crossings with low pedestrian and vehicle demand, but it causes significant interruptions to traffic flow with frequent pedestrian crossing requests. Therefore, we propose an adaptive midblock crossing control (AMCC) that minimizes the impact of the pushbutton on traffic flow while maintaining a reasonably short pedestrian wait time (PWT). We regard the midblock crossing and two adjacent intersections as an integrated system and propose two types of AMCCs-AMCC-band and AMCC-vehicle-based on different types of real-time information. AMCC-band seeks the best PWT at the midblock crossing to minimize the green band loss with downstream intersections using the signal control status of adjacent intersections. Alternatively, AMCC-vehicle leverages real-time vehicle location information [e.g., obtained from vehicle-to-infrastructure (V21) communication, connected vehicles (CVs), or advanced sensors] to minimize the estimated number of affected vehicles. Our study tests AMCC in the software Simulation of Urban MObility (SUMO) with a two-intersection traffic network. Results show that using AMCC at a midblock crossing significantly reduces vehicle delay under a wide range of traffic conditions compared to using a fixed phase and timing (Fixed) control or a pedestrian light-controlled (Pelican) crossing. The average pedestrian delay of AMCC is slightly above Pelican but much lower than Fixed. In addition, the two types of AMCCs work equally well in reducing vehicle delay, but the AMCC-vehicle has a considerably lower pedestrian delay. The results demonstrate the advantages of AMCC in reducing vehicle and pedestrian delay and vehicle stops, improving traffic efficiency at the arterial. Furthermore, the sensitivity analysis shows that the AMCC approach is adaptive to a broad range of traffic demands. Our method extends the application scope of common pushbutton control methods. We conclude that AMCC contributes to a more traffic-efficient, more pedestrian-friendly, and safer transportation system. (C) 2022 American Society of Civil Engineers.