Search

Short-time traffic flow prediction can not only help urban traffic management to complete the control and induce but also reduce the degree of urban congestion and improve the efficiency of urban operation. At the same time, improving the effect of short-time traffic flow prediction is one of the key points of traffic control and guidance development. Currently, there are many different models for short-term traffic flow prediction, and different prediction models have their different advantages and disadvantages. In this paper, we review the current status of research on short-time traffic flow prediction based on Long Short-Term Memory (LSTM) neural network: first, we analyze the research method in this paper; second, we briefly review the research and application effects of common prediction methods in traffic flow prediction; finally, the optimization method based on LSTM and its application effect are summarized and analyzed from the aspect of LSTM network and its combination model. In addition, short-time traffic flow prediction still faces challenges due to the problems of the data itself and the complexity of real-world traffic flow impact factors. [ABSTRACT FROM AUTHOR]

One of the focal points in the field of intelligent transportation is the intelligent control of traffic signals (TS), aimed at enhancing the efficiency of urban road networks through specific algorithms. Deep Reinforcement Learning (DRL) algorithms have become mainstream, yet they suffer from inefficient training sample selection, leading to slow convergence. Additionally, enhancing model robustness is crucial for adapting to diverse traffic conditions. Hence, this paper proposes an enhanced method for traffic signal control (TSC) based on DRL. This approach utilizes dueling network and double q-learning to alleviate the overestimation issue of DRL. Additionally, it introduces a priority sampling mechanism to enhance the utilization efficiency of samples in memory. Moreover, noise parameters are integrated into the neural network model during training to bolster its robustness. By representing high-dimensional real-time traffic information as matrices, and employing a phase-cycled action space to guide the decision-making of intelligent agents. Additionally, utilizing a reward function that closely mirrors real-world scenarios to guide model training. Experimental results demonstrate faster convergence and optimal performance in metrics such as queue length and waiting time. Testing experiments further validate the method's robustness across different traffic flow scenarios. [ABSTRACT FROM AUTHOR]

With the promotion of ChatGPT to the public, Large language models indeed showcase remarkable common sense, reasoning, and planning skills, frequently providing insightful guidance. These capabilities hold significant promise for their application in urban traffic management and control. However, large language models (LLMs) struggle with addressing traffic issues, especially processing numerical data and interacting with simulations, limiting their potential in solving traffic-related challenges. In parallel, specialized traffic foundation models exist but are typically designed for specific tasks with limited input-output interactions. Combining these models with LLMs presents an opportunity to enhance their capacity for tackling complex traffic-related problems and providing insightful suggestions. To bridge this gap, we present TrafficGPT—a fusion of multiple LLMs and traffic foundation models. This integration yields the following key enhancements: 1) empowering LLMs with the capacity to view, analyze, process traffic data, and provide insightful decision support for urban transportation system management; 2) facilitating the intelligent deconstruction of broad and complex tasks and sequential utilization of traffic foundation models for their gradual completion; 3) aiding human decision-making in traffic control through natural language dialogues; and 4) enabling interactive feedback and solicitation of revised outcomes. By seamlessly intertwining large language model and traffic expertise, TrafficGPT not only advances traffic management but also offers a novel approach to leveraging AI capabilities in this domain. The TrafficGPT demo can be found in https://github.com/lijlansg/TrafficGPT.git. • We propose the TrafficGPT framework, which combines large language models with expertise in traffic management. • TrafficGPT empowers LLMs to observe, analyze, process traffic data, and offer insightful decisions for traffic management. • We compare the proposed model with ChatGPT-4 Data Analyst to highlight its response accuracy and efficiency. • The adaptability and stability of TrafficGPT across various tasks, utilizing different foundational LLMs, are validated. [ABSTRACT FROM AUTHOR]

Trajectory data from connected vehicles (CVs) provide a continuous and reliable means of obtaining information that can be leveraged to optimize traffic signals. This paper proposes a real‐time traffic signal control method using CV trajectory data as the sole input. The primary goal of the proposed signal control method is to prevent queue spillover, which may significantly decrease the traffic efficiency on urban networks and induce high delays to the travelers. The proposed method formulates the signal control problem via a linear quadratic optimization model, considering the constraints related to the duration and variability of green lights in practical traffic signal control systems. Compared to conventional max‐pressure‐based methods, the optimization model offers enhanced efficiency in handling these constraints, making it highly suitable for real‐life implementation. The proposed method has undergone testing in both simulated environments and real‐world applications. In the simulation experiments, the proposed method has been demonstrated to effectively reduce spillover risks and outperform a conventional max‐pressure‐based approach even when the CV penetration rate is as low as 5%. In the real‐world experiment, the proposed method had been tested in a traffic network around the Beijing Capital International Airport for several months. The severity of spillovers, which was represented by two performance indicators, had been significantly reduced after implementing the proposed method. [ABSTRACT FROM AUTHOR]

Urbanization has led to accelerated traffic congestion, posing a significant obstacle to urban development. Traditional traffic signal scheduling methods are often inefficient and cumbersome, resulting in unnecessary waiting times for vehicles and pedestrians, exacerbating the traffic situation. To address this issue, this article proposes a dynamic traffic signal scheduling system based on an improved greedy algorithm. Unlike conventional approaches, we introduce a reward function and a cost model to ensure fair scheduling plans. A constraint function is also established, and the traffic signal scheduling is iterated through the feasible matrix using the greedy algorithm to simplify the decision-making process and enhance solution efficiency. Moreover, an emergency module is integrated to prioritize special emergency vehicles, reducing their response time during emergencies. To validate the effectiveness of our dynamic traffic signal scheduling system, we conducted simulation experiments using the Simulation of Urban Mobility (SUMO) traffic simulation suite and the SUMO traffic control interface Traci. The results indicate that our system significantly improves intersection throughput and adapts well to various traffic conditions, effectively resolving urban traffic congestion while ensuring fair scheduling plans. [ABSTRACT FROM AUTHOR]

Ensuring the correctness of road network data is critical for navigation, traffic control and urban planning. Errors like missing roads and absent connections can hinder its quality. Trajectory data emerges as a cost-effective source to uncover such errors. Existing methods often analyze the mismatches between trajectories and road networks to identify specific errors. They heavily rely on manually established rules and fail to fully leverage the diverse patterns of trajectories and the underlying road network structure. The article introduces a sequential classification approach to detect diverse road network errors. It starts with partial map matching (PMM) to associate trajectories with a road network, allowing unmatched portions. Context features are subsequently extracted by encoding patterns in the map matching (MM) outputs, raw trajectories and road network. Finally, a bidirectional recurrent neural network (BiRNN) model is trained to identify the network error category for each trajectory point. Experiments were performed on detecting errors in OpenStreetMap (OSM) road network with a real-world trajectory dataset. It demonstrates that the proposed method achieves accuracy over 96%, significantly surpassing four baselines. An ablation study confirms the necessity of considering different types of context features. This method advances error detection by effectively utilizing trajectories in identifying diverse network errors. [ABSTRACT FROM AUTHOR]

This paper presents a novel approach to intersection improvement planning utilizing telematics data from connected vehicles to generate performance measures for mobility, safety, and emissions. Congestion, crashes, and emissions are three major issues in urban areas, particularly at intersections, and agencies often struggle to prioritize improvement plans because of a lack of objective data. Traditional infrastructure sensors provide limited information at selected locations, but it is not feasible to deploy them at all intersections. The use of telematics data from connected vehicles provides a high granularity of information on driving events and trajectories that can be used in conjunction with vehicle emission modeling to efficiently generate performance measures for all intersections. In a case study of over 300 intersections in Arlington, Texas, the Pareto front method was used to evaluate and rank intersections based on multiple criteria. Intersections falling on the Pareto front were identified as having at least one outstanding (poor) performance measure and were required to be given priority for improvement. The results were cross-validated with historical crash reports and the judgments of city traffic engineers, demonstrating the effectiveness of the proposed framework in generating objective and reliable intersection performance measures. This approach has the potential to significantly improve intersection safety, mobility, and environmental impact, and can serve as a valuable decision-support tool for transportation agencies. [ABSTRACT FROM AUTHOR]

In the domain of complex urban traffic networks, real-time Traffic Signal Control (TSC) serves as a pivotal strategy for mitigating congestion. Coordinating signal control across multiple intersections involves considerable complexity. Deep Reinforcement Learning (DRL) has emerged as a robust solution. In recent years, there has been rapid advancement in TSC methods, with numerous researchers employing various novel DRL methodologies. Yet, existing surveys lack timeliness and universality in capturing the latest research. There is a notable gap in current research surveys with respect to the latest developments in TSC. Therefore, the focus of this paper lies in analyzing the most recent papers from the past five years, with the aim to provide a comprehensive and multi-dimensional review of the evolution of DRL in TSC. The survey categorizes current research based on model setups, utilized algorithms, and application scenarios. Finally, this paper highlights potential directions for future TSC research. [ABSTRACT FROM AUTHOR]

Urban traffic control (UTC) is inherently multiscale in both temporal and spatial domains. With the wide deployment of connected and automated vehicles (CAVs), data is increasingly available to help reveal this multiscale nature of UTC and the inter-dynamics among different scales. This paper applies the multiscale UTC framework we proposed earlier and extends it to UTC on a network of traffic signals with a mixed flow of CAVs and human-driven vehicles (HDVs). We adopt distributed control as the basic scheme for network-wide control and use information sharing to achieve cooperation among different intersections. We use CAV information to estimate HDV state and develop a "safety check" technique to control CAVs in the mixed traffic flow. Together, we propose a network-wide UTC framework with mixed traffic flow. To address the computation issue of the model-based multiscale method, we develop an imitation learning (IL) enhanced data-driven method to improve the computation efficiency. Specifically, we use a convolutional neural network (CNN) to represent the policies of the slower-scale signal control problem and use the data aggregation method as the learning framework to improve the policies. This algorithm's unique feature is that IL policies' training is based on the optimized results from the model-based multiscale control method. We test the model-based and IL-based methods in simulation, under various traffic scenarios and on multiple networks. We also test the transferability property of the IL-based method by training individual intersection control separately in small networks and applying them to larger networks with various types of intersections. [ABSTRACT FROM AUTHOR]

Traditional traffic signal control is one of the most widely used urban traffic control methods, due to its high efficiency, robustness, and simplicity. However, in the era of connected and automated vehicles (CAVs), it is prospective to enhance traffic control precision and achieve vehicle coordination in a more efficient manner. Focusing on an isolated intersection, this study proposes an innovative concept of Fine-grained Phase (FP) for CAV-enabled traffic control, which realizes more refined right-of-way allocation than traditional signal phases while preserving the simplicity and scalability. The development of FP jointly optimizes traffic control strategy and intersection layout design to fully utilize time-space resources, and multiple forms of FP could be generated in response to various traffic demands; meanwhile, vehicle coordination is achieved in an offline–online manner to balance solution quality and efficiency. Under the FP-based control, each lane is assigned with periodical entry time points using FP design results, then vehicles are supposed to catch up with the appropriate time points to fulfill their trips. Theoretical analyses are conducted under stationary vehicle arrival patterns, where the average delay and the intersection capacity associated with different forms of FP can be analytically acquired. For the implementation of FP-based control, we construct a general offline–online framework, consisting of an offline model that optimizes the geometry of the conflict zone at intersections and the control strategy of vehicles simultaneously, which is formulated as a mixed-integer linear programming problem, and an online FP adoption mechanism that can adapt to varying demand patterns. Under the framework, limited online computation is required, facilitating its practical real-time implementation. Simulation studies under a variety of demand settings show that FP-based control is advantageous over benchmark methods in reducing vehicle delay and also improving intersection capacity in most cases; even for cases with extremely imbalanced demands, FP-based control can admit comparable throughput, while the average delay is much smaller. In addition, results of computation time reveal that FP-based control could realize real-time implementation with negligible computation burden. • An innovative concept called FP for intersection traffic control with CAVs. • Joint optimization of traffic control and intersection layout design. • Offline–online implementation framework to balance solution quality and efficiency. • Theoretical analysis for average vehicle delay and admissible traffic capacity. • Improvement over signalized control and the state-of-art reservation strategy. [ABSTRACT FROM AUTHOR]

Large-scale congestion can lead to traffic paralysis, which severely hampers the flow of vehicles and disrupts the normal functioning of urban traffic. Traffic optimization strategies can effectively improve the performance of road networks, but often ignore the impact of regional traffic conditions and equity. This paper presents a novel traffic strategy to solve regional traffic congestion in large cities, particularly focusing on mixed traffic scenarios of connected and non-connected vehicles. The proposed method involves monitoring the traffic condition of the congestion warning community and adjusting the internal access flow within each region. The problem is formulated as a Stackelberg game, with traffic policymakers and road users as the key players. The upper layer aims to control traffic access by issuing a community warning index, with the objective of minimizing congestion warning conditions within the community. This information is then disseminated to connected vehicles which utilize it to generate personalized route guidance, while non-connected vehicles remain unaffected. The lower-level objective is to allocate vehicles in the transportation network in a user-optimal manner. To solve the bi-level programming model, the paper introduces a variable neighborhood search approach based on graph theory. The Frank-Wolfe algorithm is used to solve the lower-level model, with a penalty function introduced to transform the constrained traffic assignment problem (TAP) into an unconstrained TAP. The proposed method is applied using the data of Beijing urban road network and a sensitivity analysis is conducted to examine the impacts of critical parameters, such as regional partitioning and mixed traffic proportion. The results show that the method exhibits improved optimization performance across different parameter settings, effectively utilizing idle links and contributing to a reduction in the occurrence of traffic warning regions. • The regional control strategy prevents the emergence of warning communities. • A bi-level model is developed to solve the traffic diversion of mixed traffic. • A variable neighborhood search algorithm is designed to solve the model. • The control strategies under different critical variables scenarios are analyzed. [ABSTRACT FROM AUTHOR]

Intelligent Transportation Systems (ITS) face the formidable challenge of large-scale missing data, particularly in the imputation of traffic data. Existing studies have mainly relied on modeling network-level spatiotemporal correlations to address this issue. However, these methods often overlook the rich semantic information (e.g., road infrastructure, sensor location, etc.) inherent in road networks when capturing network-wide spatiotemporal correlations. We address this limitation by presenting the Graph Transformer-based Traffic Data Imputation (GT-TDI) model, which imputes missing values in extensive traffic data by leveraging spatiotemporal semantic understanding of road networks. The proposed model leverages semantic descriptions that capture the spatial and temporal dynamics of traffic across road networks, enhancing its capacity to infer comprehensive spatiotemporal relationships. Moreover, to augment the model's capabilities, we employ a Large Language Model (LLM) and prompt engineering to enable natural and intuitive interactions with the traffic data imputation system, allowing users to query and request in plain language, without requiring expert knowledge or complex mathematical models. The proposed model, GT-TDI, utilizes Graph Neural Networks (GNN) and Transformer architectures to perform large-scale traffic data imputation using deficient observations, sensor social connectivity, and semantic descriptions as inputs. We evaluate the GT-TDI model on the PeMS freeway dataset and benchmark it against cutting-edge models. The experimental evidence demonstrates that GT-TDI surpasses the cutting-edge approaches in scenarios with intricate patterns and varying rates of missing data. • A novel semantic description for traffic data imputation that captures spatiotemporal correlations flexibly. • A pioneering application of prompt engineering to ITS that bridges the gap between LLMs and ITS. • The proposed model leverages LLMs and semantic prompts for traffic data imputation. • Extensive experiments on real-world traffic data show the superiority. • A potential framework for integrating LLMs and semantic prompts into other ITS tasks. [ABSTRACT FROM AUTHOR]