Traffic Flow Control Research Articles

Comparison of Reinforcement Learning Agents Applied to Traffic Signal Optimisation

Traditional methods for traffic signal control at an urban intersection are not effective in controlling traffic flow for dynamic traffic demand which leads to negative environmental, psychological and financial impacts for all parties involved. Urban traffic management is a complex problem with multiple factors effecting the control of traffic flow. With recent advancements in machine learning (ML), especially reinforcement learning (RL), there is potential to solve this problem. The idea is to allow an agent to learn optimal behaviour to maximise specific metrics through trial and error. In this paper we apply two RL algorithms, one policy-based, the other value-based, to solve this problem in simulation. For the simulation, we use an open-source traffic simulator, Simulation of Urban MObility (SUMO), packaged as an OpenAI Gym environment. We trained the agents on different traffic patterns on a simulated intersection. We compare the performance of the resultant policies to traditional approaches such as the Webster and vehicle actuated (VA) methods. We also examine and contrast the policies learned by the RL agents and evaluate how well they generalise to different traffic patterns.

Microscopic state evolution model of mixed traffic flow based on potential field theory

To investigate the microscopic-state evolution mechanism of mixed traffic flow composed of Connected and Automated Vehicles (CAVs) and Human Driven Vehicles (HDVs) in the intelligent and connected environment, this paper proposed a safety potential field-NaSch (SPF-NS) model for mixed traffic flow based on potential field theory. The model introduces the safety potential field theory into the cellular automata model, reformulates the cellular automata rules and realizes the discretization of the potential field. In addition, a comparative numerical simulation experiment between NaSch model and SPF-NS model was designed. The results show that SPF-NS model can realize the real-time change of vehicle’s acceleration according to the potential field distribution, which is more precise for the actual car-following state description than NaSch model. At the same time, the road traffic flow stability under SPF-NS model is higher and the traffic capacity is increased. Besides, the microscopic-state evolution process of mixed traffic flow is emphatically studied, and the simulation experiments of different CAV penetration conditions are designed. The simulation results show that the stability of mixed traffic flow increases with the increase of CAV penetration. Compared with the HDV environment, when the CAV penetration of mixed traffic flow reaches 100%, The maximum traffic capacity of the road was increased by 2.2 times, and the congestion ratio was reduced by 96.60%. Therefore, this model can reflect the driving risk faced by vehicles in the process of car-following and simulate the microscopic-state evolution process of mixed traffic flow. The research results can provide theoretical support for future research on vehicle lane changing behavior, mixed traffic flow management and control, macroscopic state prediction in traffic flow and so on.

Traffic flow control using multi-agent reinforcement learning

One of the technologies based on information technology used today is the VANET network used for inter-road communication. Today, many developed countries use this technology to optimize travel times, queue lengths, number of vehicle stops, and overall traffic network efficiency. In this research, we investigate the critical and necessary factors to increase the quality of VANET networks. This paper focuses on increasing the quality of service using multi-agent learning methods. The innovation of this study is using artificial intelligence to improve the network’s quality of service, which uses a mechanism and algorithm to find the optimal behavior of agents in the VANET. The result indicates that the proposed method is more optimal in the evaluation criteria of packet delivery ratio (PDR), transaction success rate, phase duration, and queue length than the previous ones. According to the evaluation criteria, TSR 6.342%, PDR 9.105%, QL 7.143%, and PD 6.783% are more efficient than previous works.

Influence of Spatial Placement of Variable Speed Limit Zones on Urban Motorway Traffic Control

Traffic control approaches, in particular Variable Speed Limit (VSL), are often studied as solutions to im-prove the level of service on urban motorways. How-ever, the efficiency of VSL strongly depends on the spa-tiotemporal arrangement of VSL zones. It is crucial to determine the length and location of VSL zones for best VSL efficiency before deployment in a real system, as the optimal length of the VSL zone and their distance from the bottleneck directly affects traffic dynamics and, thus, bottleneck control. Therefore, in this study, we perform the analysis of different VSL zones lengths and their po-sitions by using a closed-loop Simple Proportional Speed Controller for VSL (SPSC-VSL). We evaluate the different VSL zone configurations and their impact on traffic flow control and vehicle emissions in a SUMO microscopic simulation on a high traffic demand scenario. The results support the observations of previous researchers on the significant dependence of VSL zone placement on VSL efficiency. Additionally, new data-based (traffic parame-ters and vehicle emissions) evidence of the performance of the SPSC-VSL design are provided regarding the best placement of consecutive VSL zones for motorway bottle-neck control not analysed in previous research.

Railway Line Occupancy Control Based on Distance Determination Sound Method.

The purpose of this research paper is to present the application of the developed sound method as a supporting tool to deal with railway traffic flow control. It is found that controlling railway line occupancy is the main issue associated with railway traffic flow. For this purpose, the line occupancy control based on a sound method has been developed. The concept of using sound waves as a source of information about approaching people, animals, vehicles, etc., has been known for centuries, and is due to the natural properties of the sense of hearing. There are many engineering attempts on the use of this phenomenon, which are mostly based on applications of distributed fiber-optic sensing technology. This paper presents the results of the sound pressure measurement in the immediate proximity of the rail to analyze and evaluate the use of the acoustic wave as an information carrier on approaching rail vehicles. The purpose of this research is to discuss the sound method introduced here, and apply it in different circumstances.

Traffic Analysis and Vehicle Flow Control using Deep Learning

In recent times, traffic congestion is a main problem at every junction.Traffic timer controlling system will display green signal for each lane at regular time intervals.If a lane does not consists of vehicles,the traffic controlling system still display the green signal for a period of time.At that time, vehicles at other lanes has to wait for the signal.This problem is prevented by the project “ TRAFFIC ANALYSIS AND FLOW CONTROL USING YOLO-v3. “YOLO-v3 stands for “You Look Only Once” and it is a real-time object detection algorithm that identifies objects in videos, images and helpful for detect object in environment. This project is developed for regulating the traffic flow by releasing better signal at time intervals keyword: Traffic flow, YOLO-v3, Deep Learning

Urban Traffic Flow Prediction with Deep Neural Network

It is critical to realize accurate collecting, visualization, rule mining, and prediction analysis of the traffic flow operating state in order for the intelligent transportation system to achieve exact management and control of traffic flow. Traffic flow prediction is primarily concerned with traffic data on roadways, which has both temporal and spatial correlations. Aiming at the spatiotemporal characteristics, this paper studies two aspects and designs a traffic flow prediction model with a deep neural network. First, this work proposes a traffic flow spatial feature learning algorithm with the combination of graph convolutional neural network and attention mechanism. Distinct weights are assigned to the degree of mutual impact between different nodes, and node adaptive learning is implemented at the same time, which modifies the standard parameter sharing mode, allowing for improved expressive ability and spatial feature extraction. Secondly, a learning algorithm for temporal characteristics of traffic flow based on the temporal convolutional network is proposed, which ensures that the dimensions of input and output data are consistent through causal convolution. The dilated convolution can flexibly control the receptive field by setting the sampling interval and can also extract temporal features well for long-length spatiotemporal sequence data. Finally, a spatiotemporal graph attention-based traffic flow prediction approach is constructed. To learn features, learn parameters for multiple modes and improve the model effect, this model employs a combination of graph convolutional neural networks and an attention mechanism. It uses a temporal convolutional network to expand the receptive field, better capture temporal features, and finally add residual connections to prevent problems such as overfitting caused by too deep network layers.

Map matching for circular road via contextual voting

Due to the limitation of positioning devices, there is a certain error between GPS positioning data and the real location on the map, and the positioning data needs to be processed to have better usability. For example, accurate location is needed for traffic flow control, automatic driving navigation, logistics tracking, etc. There are few studies specifically for circular road sections. In addition, many existing map matching methods based on Hidden Markov model (HMM) also have the problem that GPS points are easily to be matched to tangent or non-adjacent road sections at circular road sections. Therefore, the contextual voting map matching method for circular road sections (STDV-matching) is proposed. The method proposes multiple subsequent point direction analysis methods based on STD-matching to determine entry into the circular section, and adds candidate section frequency voting analysis to reduce matching errors. The effectiveness of the proposed method is verified at the circular section by comparing it with three existing HMM methods through experiments using two real map and trajectory datasets.

Modeling HDV and CAV Mixed Traffic Flow on a Foggy Two-Lane Highway with Cellular Automata and Game Theory Model

Mixed traffic composed of human-driven vehicles (HDVs) and CAVs will exist for an extended period before connected and autonomous vehicles (CAVs) are fully employed on the road. There is a consensus that dense fog can cause serious traffic accidents and reduce traffic efficiency. In order to enhance the safety, mobility, and efficiency of highway networks in adverse weather conditions, it is necessary to explore the characteristics of mixed traffic. Therefore, we develop a novel cellular automata model for mixed traffic considering the limited visual distance and exploring the influence of visibility levels and CAV market penetration on traffic efficiency. We design acceleration, deceleration, and randomization rules for different car-following scenes. For lane-changing, considering the interaction of CAVs and surrounding vehicles, we introduce game theory (GT) to lane-changing policies for CAVs. This paper presents the following main findings. In reduced visibility conditions, the introduction of CAVs is beneficial to improve mixed traffic efficiency on metrics such as free-flow speed and traffic capacity (e.g., 100% CAVs could increase the traffic capacity up to around 182% in environments of dense fog). In addition, the critical density increases as the proportion of CAVs increases, which is more pronounced in conditions of dense fog according to the simulation results. In addition, we compared the proposed GT-based lane-changing strategy to the traditional STCA lane-changing strategy. The results showed that the average speed is significantly improved under the proposed lane-changing strategy. The model presented in this paper can evaluate the overall performance and provide a reference for future management and control of mixed traffic flow in fog conditions.

Traffic flow control on cascaded roads by event‐triggered output feedback

AbstractIn this article, we propose an event‐triggered output‐feedback controller that guarantees the simultaneous stabilization of traffic flow on two connected roads. The density and velocity traffic dynamics are described with the linearized Aw‐Rascle‐Zhang macroscopic traffic partial differential equation model, which results in a coupled hyperbolic system. The control objective is to simultaneously stabilize the upstream and downstream traffic to a given spatially uniform constant steady‐state that is in the congested regime. To suppress stop‐and‐go traffic oscillations on the cascaded roads, we consider a ramp metering strategy that regulates the traffic flow rate entering from the on‐ramp to the mainline freeway. The ramp metering is located at the outlet with only boundary measurements of flow rate and velocity. Under the event‐triggered scheme, the control signal is only updated when an event triggering condition is satisfied. Compared with the continuous input signal, the event‐triggered boundary output control presents a more realistic setting to implement by ramp metering on a digital platform. The event‐triggered control design relies on the emulation of the backstepping boundary output feedback and on a dynamic event‐triggered strategy to determine the time instants at which the control value must be updated. We prove that there is a uniform minimal dwell‐time (independent of initial conditions), thus avoiding the Zeno phenomenon. We guarantee the exponential convergence of the closed‐loop system under the proposed event‐triggered control. A numerical example illustrates the results.

Detecting spatiotemporal traffic events using geosocial media data

Social media platforms enable efficient traffic event detection by allowing users to produce geo-tagged content (e.g., tweets) known as geosocial media data. Geosocial media data improve road safety by providing timely updates for traffic flow and traffic control. Recent studies on traffic event detection with geosocial media data have been focused around keyword-based query approaches, where the event content was inferred by predetermined categories, to retrieve relevant traffic events. Spatiotemporal features associated with traffic-related posts have not been fully investigated. In this study, we filtered irrelevant posts with association rules. A spatiotemporal clustering-based method was then used to retrieve traffic events from these filtered posts, where the content of detected events was automatically inferred with a set of representative terms. For comparison, a typical text classification-based method was also used by classifying the posts filtered from association rules into different categories. By validating the detection results with vehicle travel speed data, we demonstrate that the former outperforms the latter in terms of the number of correctly detected traffic events from one-year of Twitter data in Toronto, Canada. Our proposed approach helps organizations and governments to be aware of when and where traffic events occur by identifying event hotspots and peak periods, which improves both traffic management and urban planning.

A Feedback Control Method with Connected Vehicles in a Lattice Hydrodynamic Model at Highway On-Ramps

This paper presents a traffic flow control scheme of connected vehicles to stabilize a traffic system with on-ramps from a macromodel point of view. Firstly, this paper establishes a lattice hydrodynamic model with on-ramps, and an output feedback controller is designed based on the characteristics of connected vehicles. Then, by using the Lyapunov–Krasovskill functional analysis method, this paper studies the delay-dependent convergence conditions of the control method. String stability is defined as the nonamplification of the downstream disturbance of a traffic flow when it propagates upstream. The influence of the on-ramps is regarded as a stochastic disturbance in this paper. Through a theoretical analysis, the control conditions that can ensure the string stability of the traffic system under the influence of on-ramp are obtained. Finally, numerical experiments are carried out to compare the traffic states of the traffic system with and without control. The results show that the proposed control can effectively suppress the instability of the traffic system.

Improving Crowd Density Estimation by Fusing Aerial Images and Radio Signals

A recent line of research focuses on crowd density estimation from RGB images for a variety of applications, for example, surveillance and traffic flow control. The performance drops dramatically for low-quality images, such as occlusion, or poor light conditions. However, people are equipped with various wireless devices, allowing the received signals to be easily collected at the base station. As such, another line of research utilizes received signals for crowd counting. Nevertheless, received signals offer only information regarding the number of people, while an accurate density map cannot be derived. As unmanned aerial vehicles (UAVs) are now treated as flying base stations and equipped with cameras, we make the first attempt to leverage both RGB images and received signals for crowd density estimation on UAVs. Specifically, we propose a novel network to effectively fuse the RGB images and received signal strength (RSS) information. Moreover, we design a new loss function that considers the uncertainty from RSS and makes the prediction consistent with the received signals. Experimental results show that the proposed method successfully helps break the limit of traditional crowd density estimation methods and achieves state-of-the-art performance. The proposed dataset is released as a public download for future research.

Automation of the coordinated road traffic control process

Purpose. To increase output of traffic network while deriving new dependencies of transport flow characteristics upon control law parameters and using them to develop an algorithm to automate a process of the transport flow coordinated regulation within towns and cities. Methodology. The aggregated simulation model, describing processes of road traffic formation within a local section of urban traffic system, is used for the analysis of dependencies to assess quality of transport flow control upon correcting conditions being parameters of traffic cycles within intersections. The abovementioned should involve a PID controller as well as a negative feedback idea to automate control of the road traffic process. In this context, control criterion is applied to determine critical flow intensity if a traffic jam occurs. The criterion makes it possible to identify changes in the road traffic nature. To determine the optimal settings of the PID controller for different stocks of the stability control system, the standard deviation of the controlled variable from the set value and the mean static control error are defined. The derived dependencies of control quality criteria upon a law of flow regulation have helped develop a new algorithm of the coordinated automated road traffic control within a local section of urban traffic section using standard control action. Simulation experiments made it possible to assess efficiency of the proposed algorithm of control compared with available ones applied in terms of different road conditions. Findings. As a part of the studies, algorithm of the coordinated automated road traffic has been developed. It helps support maximum output of road systems while monitoring changes in the traffic environment characteristics. Originality. For the first time, dependencies of assessment criteria of traffic flow control upon the parameters of a law of flow intensity regulation have been identified if maximum output of road systems is provided. The dependencies have helped substantiate optimum control criteria for different road conditions while automating a process of coordinated traffic flow regulation within local section of urban traffic network. Practical value. The proposed dependencies of assessment criteria of traffic flow control upon the parameters of a law of flow intensity regulation as well as the algorithm of automated coordinated control are the theoretical foundations to solve such an important applied scientific problem as automation of a process of urban traffic flow regulation.

Online longitudinal trajectory planning for connected and autonomous vehicles in mixed traffic flow with deep reinforcement learning approach

This manuscript presents an Adam optimization-based Deep Reinforcement Learning model for Mixed Traffic Flow control (ADRL-MTF), so as to guide Connected and Autonomous vehicle’s (CAV) longitudinal trajectory on a typical urban roadway with signal-controlled intersections. Two improvements are made when compared with prior literatures. First, the common assumptions to simplify the problem solving, such as dividing a vehicle trajectory into several segments with constant acceleration/deceleration, are avoided, to improve the modeling realism. Second, built on the efficient Adam Optimization and Deep Q-Learning, the proposed model avoids the enumeration of states and actions, and is computational efficient and suitable for real time applications. The mixed traffic flow dynamic is firstly formulated as a finite Markov decision process (MDP) model. Due to the discretization of time, space and speed, this MDP model becomes high-dimensional in state, and is very challenging to solve. We then propose a temporal difference-based deep reinforcement learning approach, with -greedy for exploration-exploitation balance. Two neural networks are developed to replace the traditional Q function and generate the targets in the Q-learning update. These two neural networks are trained by the Adam optimization algorithm, which extends stochastic gradient descent and considers the second moments of the gradients, and is thus highly computational efficient and has lower memory requirements. The proposed model is shown to reduce fuel consumption by 7.8%, which outperforms a prior benchmark model based on Monte Carlo Tree Search. The model’s runtime efficiency and stability are tested, and the sensitivity analysis is also performed.

Resilient perimeter control for hyper-congested two-region networks with MFD dynamics

Understanding the resilience of transportation networks has received considerable research attention. Nevertheless in the field of network traffic flow control, few control approaches target the mitigation from hyper-congestion, and the control objective has rarely touched the system resilience requirement which focuses on system recovering from hyper-congested state. This paper sheds light on a resilience-oriented network control. We firstly define the traffic resilience as the integral of deviation against optimal state from disturbance generation moment t0 to recovery moment tf. Then, we propose a control method under hyper-congested situations by formulating the analytical problem using a two-reservoir transportation system with parabola-shaped Macroscopic Fundamental Diagrams (MFDs), using phase diagram analysis, attraction region derivation and switched controller design. Afterwards, we evaluate the system resilience performances between two classic perimeter control schemes (constant perimeter control (CPC) and state-feedback control (SFC)) and the proposed resilient control scheme. Results show that proposed controller can ensure the system to recover from hyper-congestion to the optimal state while existing studies failed to recover. This resilience is confirmed in various case study scenarios, e.g., when the level of hyper-congestion is different. More promisingly, the proposed control shows high compatibility with the form of the MFDs, e.g., the recover can be achieved under hysteresis conditions which are common for network-level traffic dynamics. These findings will help to design an intelligent transportation system with enhanced resilience.

Bases in problems of intelligent route selection in traffic flow control

The paper defines a simplified model of dynamic structures characteristic of the representation of traffic flows during transportation by various modes of transport. It is assumed that in each node of the transport network, the cargo flow can be distributed along branches corresponding to certain possible modes of transportation. The main parameters of branching are the share of the total volume of cargo transported by this method and the time of progress along the selected branch. A linear combination of «split» volumes of cargo flow with coefficients – volume fractions and a parameter – travel time for each transport node of the logistics chain form a basic element. The possibility of serial connection of such structures into a linear network equivalent to the original one is proved. Coefficients and time parameters in simple cases are set by expert means, in more complex cases, taking into account possible ranges of values depending on specific transport infrastructures and qualitative criteria, they are calculated using machine learning.

Assessment of Microbiological Quality of Indoor Air at Different Hospital Sites of Dilla University: A Cross-Sectional Study.

Background:In both residential and hospital indoor environments, humans can be exposed to airborne microorganisms. The hospital’s indoor air may contain a large number of disease-causing agents brought in by patients, staff, students, visitors, ventilation, or the outside. Hospitalized patients are at a higher risk of infection due to confined spaces, crowdedness, and poor infection prevention practices, which can accumulate and create favorable conditions for the growth and multiplication of microorganisms. Therefore, the aim of this study was to evaluate the indoor air bacterial load in Dilla University Hospital, Southern Ethiopia.Methods:An institutional-based cross-sectional study design was used to assess the bacterial load in the indoor air at Dilla University Hospital. To determine the bacterial load, a passive air sampling technique was used. The settle plate method was used to collect data, which involved exposing Petri-dishes filled with blood agar media to the indoor air of the sampled rooms for 60 minutes.Result:A total of 72 indoor air samples were collected once a week for 2 weeks at 14-day intervals from 18 rooms in 8 wards, and samples were collected twice a day in the morning and afternoon. The mean bacterial concentrations ranged from 450 to 1585.83 CFU/m3 after 60 minutes of culture media exposure. The mean bacterial concentrations in the obstetrics, surgical, pediatric, gynecology, and medical wards exceeded WHO guidelines. A high indoor air bacterial load was found in 58 (80.6%) of the samples in this study. Gram-positive bacteria in the air were the most common 51 (71%) of the bacterial population measured in all indoor environments. Fungal growth was found in 65 (90.3%) of the samples. Temperatures (26.5°C-28.3°C) and relative humidity (61.1%-67.8%) in the rooms were both above WHO guidelines, creating favorable conditions for bacterial growth and multiplication.Conclusion:The majority of the wards at Dilla University Hospital had bacterial loads in the air that exceeded WHO guidelines. Overcrowding, high temperatures, inadequate ventilation, improper waste management, and a lack of traffic flow control mechanisms could all contribute to a high concentration of bacteria in the indoor air. To control the introduction of microorganisms by patients, students, caregivers, and visitors, it is critical to regularly monitor indoor air bacterial load and implement infection prevention and control measures.

Safe Driving Model Based on V2V Vehicle Communication

Along with the rapid development of connected vehicle communication technology, describing the vehicle following driving status becomes gradually complicated. Driver behavior, vehicle type, and road factors affect vehicle speed, and the following distance reflects variability. In this paper, a nonlinear following distance model is constructed to characterize this variability. The model is based on the full speed difference model (FVD), and introduces the headway time distance coefficient, the following vehicle type coefficient, the communication advance response parameter reflecting the driver’s personal characteristics, and the slope coefficient and curve curvature coefficient reflecting the road conditions, etc., and analyzes to obtain the stability conditions of the model. MATLAB is applied to numerical simulation experiments of the model, and the results show that the model can better describe the variability of following headway due to driver attributes, vehicle type, slope and curve in the connected vehicle scenario, thus providing a reference for traffic flow control and management.

Моделирование передвижения транспортных средств на основе макроскопической фундаментальной диаграммы транспортного потока

The article discusses the possibilities of traffic modeling for predicting changes in vehicle accumulation and analyzing the state of traffic flow based on a macroscopic fundamental diagram at the network level. The macroscopic fundamental diagram of the traffic flow is one of the most effective and active methods for predicting the conditions of the functioning of the road network in cities. For the reliability of the study, the transport data of intelligent video surveillance on the street and road network of the central part of the city were used. Jinan (China). Based on the obtained diagrams, a piecewise two-zone system is constructed that can represent all possible states of the traffic flow. The dynamics of the dependence of the transport system between incoming and outgoing flows is shown in the frame of a piecewise two-zone system. For a specific analysis of the state of the traffic flow, modeling was carried out using the MATLAB 2019a mathematical modeling complex. As a result of traffic modeling, specific analytical equations of the system dynamics are obtained and patterns of changes in the accumulation of cars in the considered zones are presented. According to the trend of calculated car accumulation curves, a proposal has been made to design a traffic management strategy in different states of the traffic flow: for those states that are within the limit of collecting attractive curves, it is possible to apply weakened control - for those states that tend to the congestion point in the diagram, it is necessary to apply strict traffic flow control.