相关参考文献
注意：仅列出部分参考文献，下载原文获取全部文献信息。
Article
Transportation Science & Technology
Romain Ducrocq et al.
Summary: In this paper, a deep reinforcement Q-learning model is proposed to optimize traffic signal control in a partially observable environment with connected vehicles. The model adjusts traffic signals in real-time to efficiently reduce traffic congestion. It introduces a new state representation and reward function, and provides network architecture and tuned hyper-parameters to handle low detection rates.
INTERNATIONAL JOURNAL OF INTELLIGENT TRANSPORTATION SYSTEMS RESEARCH
(2023)
Article
Engineering, Civil
Haifeng Lin et al.
Summary: Urban traffic congestion, especially at intersections, causes driving delays, exhaust emissions, and fuel wastage. This study presents a mathematical model for urban trunk traffic and proposes a new optimization method combining fuzzy control theory and adaptive sequencing mutation multi-objective differential evolution algorithm for traffic signal control at urban intersections. The simulation results demonstrate the effectiveness of the proposed model optimization algorithm.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Engineering, Civil
Ammar Haydari et al.
Summary: Latest technological improvements have enhanced the quality of transportation. The emergence of new data-driven approaches has opened up new research directions for control-based systems in various domains, including transportation, robotics, IoT, and power systems. This paper presents a survey of traffic control applications based on deep reinforcement learning (RL). It extensively discusses different problem formulations, RL parameters, and simulation environments for traffic signal control (TSC) applications. The survey also covers autonomous driving applications studied with deep RL models, categorizing them based on application types, control models, and algorithms studied. The paper concludes with a discussion on challenges and open questions in deep RL-based transportation applications.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Engineering, Electrical & Electronic
Azzedine Boukerche et al.
Summary: This paper investigates the application of reinforcement learning techniques for collaborative traffic signal control in a traffic road network, while considering the issue of data transmission delay. By designing new RL agents and proposing a traffic state prediction method, the performance of the proposed method surpasses previous methods and addresses the data transmission delay issue.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2022)
Article
Green & Sustainable Science & Technology
Linan Zhang et al.
Summary: A growing trend in urban-road network traffic flow is the inclusion of intelligent connected vehicles alongside regular vehicles, which requires research in cooperative control methods, traffic control application scenarios, and experimental validation. This study focuses on the characteristics and information extraction methods of heterogeneous traffic flow, as well as different driving modes for regular vehicles and intelligent connected vehicles. The research also examines control modes such as traffic-signal control, active control of intelligent connected vehicles, and indirect control of regular vehicles through intelligent vehicles. The study proposes an active control mechanism and multi-intersection coordinated control strategy for intelligent connected vehicle heterogeneous traffic flow, combining coordinated control theory and conducting integrated experiments.
Article
Mathematics
Vladimir Shepelev et al.
Summary: This study proposes a method to predict the release time of vehicles passing controlled intersections using neural networks and object trackers, reducing stops by optimizing speed modes. The technique allows for transmitting predicted delay times through V2X communication locally and through HAV-the Digital Road Model in horizontally scaled mode.
Article
Mathematics
Rifkat Minnikhanov et al.
Summary: This paper proposes and evaluates an approach for detecting vehicle trajectory abnormalities in CCTV videos from road intersections. The focus is on trajectory analysis rather than object detection and tracking. The suggested approach overcomes spatial perspective challenges and improves performance through trajectory approximation and thinning techniques. Modification of the similarity metric is suggested to consider spatial perspective. Clustering is used to identify normal and abnormal trajectory classes. A framework implementing the approach is developed and experiments are conducted to test and provide recommendations for using different techniques within its scope.
Article
Mathematics
Yafeng Gu et al.
Summary: In this paper, the authors propose a spatial-temporal attention graph convolution network (STAGCN) to address the challenges of capturing global and local spatiotemporal dependencies in traffic data. By acquiring static and dynamic graphs, and introducing a gated temporal attention module, the method achieves outstanding accuracy improvement in traffic flow prediction.
Article
Transportation Science & Technology
Qinzheng Wang et al.
Summary: This study proposes a decentralized approach to optimize the trajectories of automated vehicles, aiming to alleviate traffic congestion and improve traffic operation efficiency. A two-stage model is developed to optimize the trajectories of vehicles traveling along a road, with the objective of minimizing delay and lane-changing costs.
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
(2022)
Article
Engineering, Civil
Dongfang Ma et al.
Summary: In this paper, a novel deep actor-critic method is proposed for efficient traffic signal control, combining deep neural networks and actor-critic models. Simulation experiments demonstrate that this method outperforms classic model-based approaches and existing deep reinforcement learning methods across various performance measures.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Engineering, Civil
Mehrdad Tajalli et al.
Summary: This study introduces a methodology for cooperative signal timing and trajectory optimization at intersections with a mix of CAVs and HVs, using a mixed-integer non-linear program to find near-optimal solutions efficiently in complex intersections. By linearizing nonlinear constraints and decomposing the program, the methodology provides a balance between computational efficiency and solution quality, resulting in reduced travel time and fuel consumption under different scenarios.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Computer Science, Information Systems
Maanak Gupta et al.
Summary: ITS aims to provide safe, secure and smart travel experience through V2V and V2I communication, but privacy and security concerns hinder its adoption. A secure V2V and V2I communication approach using cloudlet support is proposed to ensure message authenticity and anonymity, with performance metrics compared to existing technologies.
IEEE TRANSACTIONS ON SERVICES COMPUTING
(2022)
Article
Mathematics
Sadiqa Jafari et al.
Summary: Traffic congestion is a significant issue, and the suggested method proposes the use of fuzzy logic control to reduce traffic volume, waiting times, and collisions. The method is investigated using a multi-agent system at intersections and the stability of each intersection is proven. Simulation results show that fuzzy modeling is effective in reducing queueing times for vehicles.
Article
Computer Science, Artificial Intelligence
Yan Wang et al.
IEEE Transactions on Intelligent Vehicles
(2022)
Article
Mathematics
Shihao Zhao et al.
Summary: This paper proposes an attention and wavelet-based spatial-temporal graph neural network for traffic flow and speed prediction. By integrating attention mechanism and graph wavelet neural network, it captures local and global spatial information, and extracts time series information using a gated temporal convolutional network and temporal attention mechanism.
Article
Optics
A. A. Agafonov et al.
Summary: In this study, a novel hybrid reinforcement learning-based approach was developed for solving the traffic signal control problem. The approach utilized deep learning to predict the number of vehicles at intersections and employed double Q-learning for signal control. Experimental results demonstrated that the proposed algorithm outperformed other algorithms in terms of average waiting time and average travel time.
OPTICAL MEMORY AND NEURAL NETWORKS
(2022)
Article
Physics, Multidisciplinary
A. A. Agafonov et al.
Summary: We propose a two-stage method for adaptive traffic signal control based on estimating the predicted weighted flow of vehicles passing through an intersection. The first stage involves estimating the time for each vehicle to pass the intersection using an artificial neural network model, followed by estimating the predicted flow of vehicles for a given phase of the traffic light cycle. In the second stage, the weighted flow is estimated and vehicle waiting time is considered. The method selects the traffic light phase by maximizing the weighted traffic flow, and experimental studies show its superiority over classical approaches and state-of-the-art methods based on reinforcement learning.
OPTOELECTRONICS INSTRUMENTATION AND DATA PROCESSING
(2022)
Proceedings Paper
Operations Research & Management Science
Yaohui An et al.
Summary: This paper proposes a modified proximal policy optimization algorithm to adaptively regulate traffic signals and alleviate intersection congestion. By using images as state inputs, the complexity of the state space is reduced. Experimental results show significant improvements in cumulative reward and average waiting time compared to other deep reinforcement learning algorithms, demonstrating the effectiveness of the proposed algorithm.
2022 10TH INTERNATIONAL CONFERENCE ON TRAFFIC AND LOGISTIC ENGINEERING (ICTLE 2022)
(2022)
Article
Transportation Science & Technology
Yi Guo et al.
Summary: The paper proposed a real-time learning and control framework for signalized intersection management that integrates perception, prediction, planning, and optimization components to improve traffic efficiency. This framework utilizes LSTM networks to learn traffic patterns and driver behavior, and deep reinforcement learning to solve signal optimization problems, resulting in increased throughput and decreased delay.
TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES
(2021)
Article
Engineering, Electrical & Electronic
Yu Du et al.
Summary: The paper proposes a Coupled Vehicle-Signal Control (CVSC) method to optimize traffic signal timing and driving trajectories of CAVs simultaneously, aiming at improving traffic efficiency and energy saving. The research results show that when the penetration rate of CAVs is greater than 40%, this method can effectively enhance the performance of signalized intersections by reducing fuel consumption by 6%-14% and increasing average speed by 1%-5%.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2021)
Article
Mathematics
Yaroslav Kholodov et al.
Summary: This paper introduces a generalized second-order hydrodynamic traffic model and demonstrates that well-known models like Payne-Whitham, Aw-Rascal, and Zhang are all special cases of this generalized model. The proposed model is validated using traffic data from a segment of the Interstate 580 freeway in California, showing its effectiveness in traffic modeling and analysis.
Proceedings Paper
Automation & Control Systems
Wei Miao et al.
Summary: Traffic congestion is a significant issue in urban governance, and deep reinforcement learning is being applied to traffic signal control for optimizing roadside traffic timing. This advanced artificial intelligence method shows promise in addressing traffic bottlenecks in urban areas.
PROCEEDINGS OF THE 33RD CHINESE CONTROL AND DECISION CONFERENCE (CCDC 2021)
(2021)
Article
Engineering, Civil
Tianshu Chu et al.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2020)
Article
Automation & Control Systems
Tian Tan et al.
IEEE TRANSACTIONS ON CYBERNETICS
(2020)
Article
Robotics
Xun Shen et al.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2020)
Article
Chemistry, Multidisciplinary
Dinh Dung Nguyen et al.
APPLIED SCIENCES-BASEL
(2020)
Article
Economics
Chunhui Yu et al.
TRANSPORTATION RESEARCH PART B-METHODOLOGICAL
(2019)
Article
Optics
A. A. Agafonov et al.
OPTICAL MEMORY AND NEURAL NETWORKS
(2019)
Article
Engineering, Civil
Walid Balid et al.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2018)
Article
Economics
Fang Zhou et al.
TRANSPORTATION RESEARCH PART B-METHODOLOGICAL
(2017)
Article
Economics
Jiaqi Ma et al.
TRANSPORTATION RESEARCH PART B-METHODOLOGICAL
(2017)
Article
Engineering, Multidisciplinary
Isabel M. Ribeiro et al.
ENGINEERING OPTIMIZATION
(2016)
Article
Engineering, Civil
Luis Moreira-Matias et al.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2015)
Article
Engineering, Civil
Li Li et al.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2014)
Review
Engineering, Electrical & Electronic
M Papageorgiou et al.
PROCEEDINGS OF THE IEEE
(2003)
