Traffic Control Algorithm Research Articles

Logic-Based Traffic Flow Control for Ramp Metering and Variable Speed Limits—Part 1: Controller

This paper proposes a Logic-Based Traffic Flow Control algorithm (LB-TFC) for integrated control of Ramp Metering (RM) installations and Variable Speed Limits (VSLs) in order to reduce traffic jams created at bottlenecks. LB-TFC estimates, for each control time step, the number of vehicles that should be held back or released by the control measures (i.e. the VSLs and the RM rates) in order to avoid the capacity drop (maximizing the outflow of the bottleneck). Afterwards, based on the resulting estimated number of vehicles, the VSLs and/or the RM rates are increased or decreased in a pre-specified order. In order to avoid or reduce traffic breakdowns, the proposed controller (LB-TFC) anticipates the future evolution of the bottleneck density by using a feed-forward structure. As a result, the performance of the controller is very efficient and similar to the one obtained with an optimal controller while the implementation of the controller (with an almost instantaneous computation time) and the tuning of the parameters are easy. In the second part of this work, published in a separate paper (`Part 2: Simulation and Comparison'), LB-TFC is simulated, analyzed and compared for two freeways (one synthetic network and one stretch of the ring-road freeway SE-30 in Seville, Spain).

Double Deep Q-Network with a Dual-Agent for Traffic Signal Control

Adaptive traffic signal control (ATSC) based on deep reinforcement learning (DRL) has shown promising prospects to reduce traffic congestion. Most existing methods keeping traffic signal phases fixed adopt two agent actions to match a four-phase suffering unstable performance and undesirable operation in a four-phase signalized intersection. In this paper, a Double Deep Q-Network (DDQN) with a dual-agent algorithm is proposed to obtain a stable traffic signal control policy. Specifically, two agents are denoted by two different states and shift the control of green lights to make the phase sequence fixed and control process stable. State representations and reward functions are presented by improving the observability and reducing the leaning difficulty of two agents. To enhance the feasibility and reliability of two agents in the traffic control of the four-phase signalized intersection, a network structure incorporating DDQN is proposed to map states to rewards. Experiments under Simulation of Urban Mobility (SUMO) are carried out, and results show that the proposed traffic signal control algorithm is effective in improving traffic capacity.

Traffic density estimation using progressive neural architecture search

The current research work is to develop an automated system for estimating the traffic density of roads at traffic junctions and proposed a traffic control algorithm for smooth movement of vehicles based upon the information provided by the density estimation algorithm. The existing methodologies of hand-crafted features and rule-based estimation to count the number of vehicles has been replaced here by Convolutional Neural Network (CNN). The conventional CNN has been replaced by Progressive Neural Architecture (PNAS) which is based on sequential model-based optimization (SMBO) strategy.

Link Flow Estimation on an Isolated Intersection Based on Deep Learning Models

This paper considers the problem of short-term estimation of the link flow on an isolated intersection by means of artificial intelligence (AI). The problem is formulated as a supervised learning regression task. Several architectures of neural networks are presented to predict the link flow: recurrent neural network (RNN), convolutional neural network (CNN), and a combination of the two, with RNN treating the time dependencies and CNN extracting the spatial properties from traffic. The models are trained on synthetic data that are generated by a simulator. These models are compared with a fully-connected network. Numerical experiments are carried out on CPU and GPU. They show that deep learning models with special architecture can estimate the link flow much better than a traditional neural network on a single intersection. The obtained link flow estimates can be used in different traffic control algorithms in order to minimize the transport network congestion.

Research and application of intersection traffic signal control algorithm based on vehicle location

This paper focuses on the optimising algorithms of signal timing of urban intersection combined with vehicle location using the platform of the new type of wireless electronic license plate system. With the traffic flow information obtained from the platform, a dynamic optimisation model with optimal control parameters of green light time of all four phases for single intersection is constructed. Then genetic algorithm is applied to obtain the satisfying signal timing scheme. In simulation examples, the proposed dynamic distribution method and the traditional time control method are compared. The results show that compared with the traditional time control method, in traffic rush hour, the genetic optimisation algorithm can reduce the waiting time by 23% with the same road vehicle capacity. At non-peak time, the proposed method can reduce the waiting time by 10%. Therefore, the introduced method can effectively reduce the total queue delay, and improve the traffic capacity of the intersection.

Situation-cognitive traffic light control based on object detection using YOLO algorithm

Current traffic lights provide the green signal with fixed-time interval without considering the traffic situation. As a result, cars in a long line have to wait long time, which causes traffic jams and makes the drivers be irritated. In order to solve the problem, it is necessary to control the green signal interval according to the analysed traffic volume using the image processing and the machine learning techniques. This paper presents a situation-cognitive traffic light control algorithm that measures the traffic volume using object detection algorithm called you only look once (YOLO) and controls the traffic signal intervals according to the traffic volume. The algorithm expects the smooth traffic flow and the reduction of the driver's stress.

Situation-cognitive traffic light control based on object detection using YOLO algorithm

Current traffic lights provide the green signal with fixed-time interval without considering the traffic situation. As a result, cars in a long line have to wait long time, which causes traffic jams and makes the drivers be irritated. In order to solve the problem, it is necessary to control the green signal interval according to the analysed traffic volume using the image processing and the machine learning techniques. This paper presents a situation-cognitive traffic light control algorithm that measures the traffic volume using object detection algorithm called you only look once (YOLO) and controls the traffic signal intervals according to the traffic volume. The algorithm expects the smooth traffic flow and the reduction of the driver's stress.

Self-adaptive Traffic and Logistics Flow Control using Learning Agents and Ubiquitous Sensors

Traffic flow optimisation is a distributed complex problem. Traditional traffic and logistics flow control algorithms operate on a system level and address mostly switching cycle adaptation of traffic signals and lights. This work addresses traffic flow optimisation by self-adaptive micro-level control by combining Reinforcement Learning and rule-based agent models for action selection with a new hybrid agent architecture. I.e., long-range routing is performed by agents that adapt their decision making for re-routing on local environmental sensors. Agent-based modelling and simulation are used to study emergence effects on urban city traffic flows with learning agents. The approach and the proposed agent architecture can be generalised and applied to a broader range of application fields, e.g., logistics and general transport phenomena.

A heuristic method to optimize generic signal phasing and timing plans at signalized intersections using Connected Vehicle technology

This paper develops a flexible, real-time traffic signal control algorithm to optimize both phase durations and phase sequences at four-approach intersections with conflicting left-turns, based on information obtained from connected vehicles. Vehicle-to-infrastructure communications are assumed to provide the location of all connected vehicles near the signalized intersection at regular time intervals and this information is used to identify the presence of non-connected vehicles that are stopped at the intersection. All detected and identified vehicles are used to identify naturally occurring platoons in the traffic stream. The signal control algorithm then selects the optimal sequence that these platoons should discharge through the intersection to minimize average delay of all identified vehicles. Since all possible departure sequences for platoons of vehicles are considered, the problem is computationally difficult. Hence, several heuristic methods are proposed to determine optimal platoon departure sequences. These heuristics include an intelligent tree search and multiple types of genetic algorithms, including a newly developed genetic algorithm that perserves phase order sequence that is vital to this problem. Comparisons between these heuristics and the global optimal solution suggest that the heuristics are able to provide similar operational performance with significant reductions in total computation time required, such that the algorithm can be applied in real-time. In general, the intelligent tree search appears to outperform the genetic algorithm approaches in terms of operational performance but has computational requirements that increase exponentially with the number of platoons identified at the intersection. Meanwhile, the genetic algorithm methods tend to be more scalable but slightly less efficient. Overall, the results are promising for the application of the proposed flexible signal control algorithm at real intersections.

Cooperative Traffic Signal Control with Traffic Flow Prediction in Multi-Intersection.

As traffic congestion in cities becomes serious, intelligent traffic signal control has been actively studied. Deep Q-Network (DQN), a representative deep reinforcement learning algorithm, is applied to various domains from fully-observable game environment to traffic signal control. Due to the effective performance of DQN, deep reinforcement learning has improved speeds and various DQN extensions have been introduced. However, most traffic signal control researches were performed at a single intersection, and because of the use of virtual simulators, there are limitations that do not take into account variables that affect actual traffic conditions. In this paper, we propose a cooperative traffic signal control with traffic flow prediction (TFP-CTSC) for a multi-intersection. A traffic flow prediction model predicts future traffic state and considers the variables that affect actual traffic conditions. In addition, for cooperative traffic signal control in multi-intersection, each intersection is modeled as an agent, and each agent is trained to take best action by receiving traffic states from the road environment. To deal with multi-intersection efficiently, agents share their traffic information with other adjacent intersections. In the experiment, TFP-CTSC is compared with existing traffic signal control algorithms in a 4 × 4 intersection environment. We verify our traffic flow prediction and cooperative method.

Max-pressure traffic controller based on travel times: An experimental analysis

The traffic control of an arbitrary network of signalized intersections is considered. This work presents a new version of the recently proposed max-pressure controller, also known as back-pressure. The most remarkable features of the max-pressure algorithm for traffic signal control are: scalability, stability, and distribution. The modified version presented in this paper improves the practical applicability of the max-pressure controller by considering as input travel times instead of queue lengths. The two main practical advantages of this new version are: (i) travel times are easier to estimate than queue lengths, and (ii) max-pressure controller based on travel times has an inherent capacity-aware property, i.e., it takes into account the finite capacity of each link. Travel time tends to diverge when the queue length is close to its capacity. It should be noted that previous max-pressure algorithms rely exclusively on queue length measurements, which may be difficult to accomplish in practice. Moreover, these previous algorithms generally assume queues with unbounded capacity. This may be problematic because a model with unbounded capacity links is not able to reproduce spillbacks, which are one of the most critical phenomena that a traffic signal controller should avoid. After presenting the new version of the max-pressure controller, it is analyzed and compared with existing control policies in a microscopic traffic simulator. Moreover, results of a real implementation of the developed algorithm to a signalized intersection, located at an urban arterial in Jerusalem, are shown and analyzed. To the best of the authors’ knowledge, this experiment is the first real implementation of a max-pressure controller at a signalized intersection.

An equitable traffic signal control scheme at isolated signalized intersections using Connected Vehicle technology

This paper extends a real-time, Connected Vehicle-based traffic signal control algorithm at isolated intersections to balance between two competing intersection objectives: efficiency and equity. In this approach, a central controller is used to collect real-time locations of connected vehicles at regular intervals, which can then be used to also identify the existence of some of the non-connected vehicles. The control algorithm then aims to optimize the discharge sequence of naturally occurring platoons of vehicles based on their proximity. Specifically, the strategy selects the platoon discharge sequence—and associated signal phase and timing plan—that minimizes average vehicle delay (measure of efficiency) while limiting the maximum delay any individual vehicle may experience (measure of equity). The latter objective is only possible with detailed vehicle-level information available from connected vehicles. The results show that without the threshold on maximum individual vehicle delay, average delay is often minimized at the expense of very large delays imposed onto some vehicles. By implementing a threshold, both the maximum vehicle delay and the distribution of individual vehicle delays—as measured by the standard deviation and a common measure of population inequity, the Gini coefficient—can be improved, often with only negligible impacts to intersection efficiency. However, the tradeoff between equity and efficiency becomes more significant as the maximum vehicle delay threshold decreases. Sensitivity tests show that this control algorithm works well for different total traffic demands and different demand patterns. The proposed algorithm is also effective under imperfect connected vehicle penetration rates when the connected vehicles make up more than 40% of the traffic stream. The results suggest that the proposed strategy can help significantly reduce long delays and inequitable treatment of vehicles at an intersection when vehicle-level information is available to a signal controller.

Evaluation of the effectiveness of traffic control algorithms based on a simulation model in the AnyLogic

The increase in the number of vehicles every year covers an growing number of countries, as a result of which the vehicle fleet of large cities and megalopolises is constantly increasing, as well as the number of people involved in traffic. The growth in traffic leads to the increase in traffic intensity, which in the conditions of cities with historically developed buildings leads to a transport problem. It is particularly acute in the nodal points of the road network. The problems are caused by the increase in transport delays, queuing and congestion, which in turn causes the decrease in the speed of communication, unnecessary consumption of fuel and increased wear of components and assemblies of vehicles. Effective traffic management involves the use of both network and local adaptive control algorithms. City-wide automated system should be based on the principles of maximum management efficiency and reliability of the system.

Dynamic Control of Traffic Signals using Traffic Data from Google Maps and Road Cameras

Traffic congestion is a condition that occurs due to over usage of roads and results in increased waiting time, slower speeds and increased vehicle queuing. Every year billions of dollars are lost due to time wasted by waiting at traffic signals. In this paper, the problem of dynamic control of traffic signals using traffic data obtained from Google Maps and Road Cameras is explored. The traffic signal control algorithm described will adjust both the ordering of different lights in real time as well as the duration of each of them. The ordering (sequence) and length of each traffic signal light for a period of time will be affected by factors such as traffic volume, waiting time and vehicle queuing characteristics. The results from the above algorithm’s simulations are compared to the average waiting times the vehicles experience in the case of static traffic signals which employ a fixed duration and sequence policy for each of the colors of the traffic signal.

A CASE FOR IMPLEMENTING SELF-ORGANISING TRAFFIC SIGNAL CONTROL ON SOUTH AFRICAN ROADS

Traffic signal optimisation may lead to the alleviation, to some extent, of urban traffic congestion, particularly by using real-time data rather than expected traffic flow data. Recent advances in radar technology have made it possible to observe detailed traffic flow data in and around roadway intersections in real time. The notion of self-organisation has relatively recently been proposed as a promising alternative to improve the effective allocation of green time, particularly under lighter traffic conditions. A fixed-time control strategy and seven self-organising algorithms are compared in a microscopic traffic simulation model of a provincial road in the Western Cape province of South Africa. Actual arrival rates are used as input for the model, while the algorithms are compared using six performance measure indicators, under both light and moderate traffic conditions. The results are used to make a case for the adoption of self-organising traffic signal control algorithms, especially under conditions of light to moderate traffic densities, since this can lead to significant improvements in traffic flow in terms of delay time, vehicle stops, and time spent travelling at unacceptably slow speeds through the road network.

Self-organisation in traffic signal control algorithms under light traffic conditions

Fixed-time control and vehicle-actuated control are two distinct types of traffic signal control. The latter control method involves switching traffic signals based on detected traffic flows and thus offers more flexibility (appropriate for lighter traffic conditions) than the former, which relies solely on cyclic, predetermined signal phases that are better suited for heavier traffic conditions. The notion of self-organisation has relatively recently been proposed as an alternative approach towards improving traffic signal control, particularly under light traffic conditions, due to its flexible nature and its potential to result in emergent behaviour. The effectiveness of five existing self-organising traffic signal control strategies from the literature and a fixed-control strategy are compared in this paper within a newly designed agent-based, microscopic traffic simulation model. Various shortcomings of three of these algorithms are identified and algorithmic improvements are suggested to remedy these deficiencies. The relative performance improvements resulting from these algorithmic modifications are then quantified by their implementation in the aforementioned traffic simulation model. Finally, a new self-organising algorithm is proposed that is particularly effective under lighter traffic conditions.

Uncertainty-Aware Flight Scheduling for Airport Throughput and Flight Delay Optimization

The continuous growth in the demand for air transportation exceeds the capacity of existing infrastructure, usually leading to unreliable flight schedules, i.e., long flight delays and uncertainties in arrival/departure and taxi times. We tackle the problem in this paper by designing an air traffic control algorithm, which can accommodate both airport throughput and flight quality of service in terms of flight delay on a given runway. The flight scheduling problem is formulated as an integer linear programming, and then converted to a multiobjective optimization problem which enables the computation of tradeoff between scheduling resolution and time complexity. Based on the multiobjective optimization, a heuristic algorithm considering uncertainties in flight arrival/departure time and taxi time is designed to achieve an improvement in airport throughput and a reduction in flight delay. Extensive simulations show that compared to benchmarking schemes, the proposed uncertainty-aware flight scheduling algorithm can improve the airport throughput and flight delay by up to 12.02% and 31.4%, respectively.

Framework of simulation-based vehicle safety performance assessment of highway system under hazardous driving conditions

Vehicles are extremely vulnerable to single-vehicle accidents under some hazardous driving conditions (i.e. strong wind, icy or snowy road surface). An integrated framework is proposed to assess single-vehicle traffic safety performance of stochastic traffic flow under hazardous driving conditions. Different from most existing studies focusing on a single vehicle moving at a constant speed, for the first time, the proposed work evaluates individual vehicle safety performance based on the time-dependent simulation results of stochastic traffic flow, including instantaneous speeds and positions of each vehicle as a part of simulated traffic flow. Simultaneously, complex geometric and other environmental conditions of the highway system are also considered realistically, not only during the safety assessment process, but also in quantifying the wind loads applied on the vehicles. Finally, with the safety information of each individual vehicle, an overall safety performance index of the whole traffic flow on the highway system is further introduced, which serves as a potential traffic safety performance measure and resilience indicator of transportation infrastructure systems under various hazards. This study has potential applications to not only regular vehicles, but also advanced traffic management and control algorithms for connected and autonomous vehicles in hazardous driving environments.

Dynamic traffic demand uncertainty prediction using radio‐frequency identification data and link volume data

Dynamic traffic demand is crucial for developing effective strategies and algorithms for real-time traffic management and control. The uncertainty of traffic demand provides additional information while its prediction is very complicated and is inadequately investigated in the existing literature. Recently, radio-frequency identification (RFID) technology has been deployed to monitor the condition of traffic. In this study, the authors propose a modelling system to predict dynamic traffic demand uncertainty using the RFID data and the link traffic volume. The modelling system includes an optimisation model using the RFID data to calculate the historical traffic demand in terms of origin–destination matrix, a state-space model solved by Kalman filter to predict the traffic demand, and a GARCH model to capture the conditional variance of the predicted traffic demand. Performance measures are applied to evaluate prediction accuracy. The case study in Nanjing, China shows that the modelling system has the desirable performance to successfully predict uncertainties of dynamic traffic demand.

Production logistics and human-computer interaction\u2014state-of-the-art, challenges and requirements for the future

Current concept, development, and testing applications in production concerning Cyber-Physical Systems (CPS), Industry 4.0 (I40), and Internet of Things (IoT) are mainly addressing fully autonomous systems, fostered by an increase in available technologies regarding distributed decision-making, sensors, and actuators for robotics systems. This is applied also to production logistics settings with a multitude of transport tasks, e.g., between warehousing or material supply stations and production locations within larger production sites as for example in the automotive industry. In most cases, mixed environments where automated systems and humans collaborate (e.g., cobots) are not in the center of analysis and development endeavors although the worker’s adoption and acceptance of new technologies are of crucial relevance. From an interdisciplinary research perspective, this constitutes an important research gap, as the future challenges for successful automated systems will rely mainly on human-computer interaction (HCI) in connection with an efficient collaboration between motivated workers, automated robotics, and transportation systems. We develop a HCI efficiency description in production logistics based on an interdisciplinary analysis consisting of three interdependent parts: (i) a production logistics literature review and process study, (ii) a computer science literature review and simulation study for an existing autonomous traffic control algorithm applicable to production logistics settings with the specific inclusion of human actors, and (iii) a work science analysis for automation settings referring to theoretical foundations and empirical findings regarding the management of workers in digital work settings. We conclude with practical implications and discuss avenues for future research and business applications.