In large-scale traffic optimization, models based on the Macroscopic Fundamental Diagram (MFD) are recognized for their efficiency in broad network analyses. However, they fail to reflect variations in the individual traffic status of each road link, leading to a gap in detailed traffic optimization and analysis. To address the limitation, this study introduces a Local Correction Factor (LCF) that represents local speed deviations between the actual link speed and network mean speed from the MFD based on the link configuration. The LCF is calculated using a deep learning approach that takes as inputs the network mean speed from the MFD and the road network configuration. Our framework integrates Graph Attention Networks (GATs) with Gated Recurrent Units (GRUs) to capture both the spatial configurations and temporal correlations within the network. Coupled with a strategic network partitioning method, our model enhances the precision of link-level traffic speed estimations while preserving the computational advantages of aggregate models. In our experiments, we evaluate the proposed LCF across various urban traffic scenarios, including different levels of origin-destination trip demand and distribution, as well as diverse road configurations. The results demonstrate the robust adaptability and effectiveness of the proposed model. Furthermore, we validate the practicality of our model by calculating the travel time of each randomly generated path, achieving an average error reduction of approximately 84% relative to MFD-based results.

Traffic forecasting is a fundamental task in transportation research, however, the scope of current research has mainly focused on a single data modality of loop detectors. Recently, the advances in Artificial Intelligence and drone technologies have enabled novel solutions to efficient, accurate, and flexible aerial observations of urban traffic. As a promising traffic monitoring approach, drones can create an accurate multi-sensor mobility observatory for large-scale urban networks, when combined with existing infrastructure. Therefore, this paper investigates urban traffic prediction from a novel perspective, where multiple input sources, i.e., loop detectors and drones, are utilized to predict both the segment-level traffic of all road segments and the regional traffic. A simple yet effective graph-based model, Hierarchical Multi-Source Neural Network (HiMSNet), is proposed to integrate multiple data modalities and learn spatio-temporal correlations. Detailed analysis shows that predicting accurate segment-level speed is more challenging than the regional speed, especially under high-demand scenarios with heavier congestion and varying traffic dynamics. Utilizing both drone and loop detector data, the prediction accuracy can be improved compared to single-modality cases, when the sensors have lower coverage and are subject to noise. Our simulation study based on vehicle trajectories in a real urban road network has highlighted the value of integrating drones in traffic forecasting and monitoring.

Monitoring outdoor urban parking areas has traditionally relied on either costly sensing technology, such as ground sensors, or manual inspections, which are both resource-intensive. The emergence of drones equipped with advanced visual sensors provide a comprehensive aerial perspective and versatile mobility, opening up new possibilities for efficient traffic monitoring. In this research, we demonstrate the efficient monitoring of usage levels in outdoor public parking lots using drones. We deployed a number of drones flying during the peak periods for two days over four major outdoor parking lots in Pully, Switzerland, while monitoring on-street parking in their proximity. Our proposed pipeline involves identifying parking areas through geo-referencing and an image feature matching approach, followed by vehicle detection using an innovative boosted pseudo-labeling method. Central to our pipeline is an innovative boosted pseudo-labeling technique that enhances detection accuracy by generating pseudo labels from stationary vehicles observed in multiple frames, thereby reducing the need for manual data annotations. From the video collected from our drone experiment, we automate the monitoring of the outdoor parking areas over time and days. We conduct a comprehensive analysis of the occupancy rate of each parking area, encompassing both off-street and on-street parking lots, as well as dynamic interactions between different locations, and also examined the turnover rate of individual parking spots. This research represents a significant innovation in the use of drones for parking studies, providing an effective, versatile, and insightful approach for studying urban mobility and traffic management.

Unmanned aerial systems, known as “drones,” are relatively new in collecting traffic data. Data from drone videography can have potential applications for traffic research. Drones can record the vehicles from their aerial point-of-view and provide their naturalistic driving behavior. Processing raw data from drones to remove noise and anomalies is crucial to ensure that the data are fit for subsequent applications, e.g., the development of traffic flow or crash risk models. This study uses a part of the pNEUMA dataset, a large dataset with almost half a million trajectories captured by a swarm of drones over Athens, Greece. This novel dataset offers an opportunity to analyze the data attributes and treat the noise and outliers in the data. We use a combination of smoothing filters and Extreme Gradient Boosting with adaptive regularization to process the speed and acceleration profiles of the vehicle trajectories in the dataset. Our approach can help prospective data users treat this or similar trajectory datasets alternatively to applying manual thresholds and assist in accelerating research in microscopic traffic analysis.

Perimeter control based on Macroscopic Fundamental Diagram (MFD) aims to meter the number of transferring vehicles at the periphery of the protected urban region in order to obtain the desired number of vehicles in that region. The advantage of perimeter control is less computational effort, while its drawback is that it may create long queues and delays at the perimeter of the controlled area. For capturing boundary queue dynamics, an enhanced accumulation-based MFD model is proposed using colored Petri Nets by considering transfer flows, boundary queues and travel delays simultaneously. The gated intersections and related road segments on the border of a protected region are modeled as so-called boundary buffers. Based on the enhanced MFD model, anintegrated perimeter control framework is proposed with consideration of travel time and queuing time in buffers. In this framework, the controllers between peripheral and protected region are optimized using model predictive control theory. Then, internal flow controllers are adopted to homogenize traffic density among subregions, and route guidance is also used to balance the number of queuing vehicles among boundary buffers. Simulation results verify the effectiveness of the proposed integrated perimeter control. Furthermore, the impacts of buffer storage capacity on region heterogeneity and trip completion rates are also investigated in this paper.Note to Practitioners—It is challenging to manage traffic congestion in large-scale urban network. Perimeter control provides an accumulation-based methodology with consideration of the existing correlation between traffic density and flow, which is known as MFD. For practical application, the efficiency as well asweakness of potential perimeter control strategies need to be evaluated and improved using customized traffic simulations. Accumulation-based traffic model using Petri Nets is introduced to serve for perimeter control, in which the intersections and road segments on the boundary of each pair of adjacent subregions are modeled as a boundary buffer. Both perimeter control and route guidance are integrated in the proposed control framework considering the queuing vehicles in the boundary buffers. Moreover, the effect of buffer storage capacity on network performances is tested, which is the essential for traffic engineers to design and implement management measures in practice.

Accurate prediction of travel time is an essential feature to support Intelligent Transportation Systems (ITS). The non-linearity of traffic states, however, makes this prediction a challenging task. Here we propose to use dynamic linear models (DLMs) to approximate the non-linear traffic states. Unlike a static linear regression model, the DLMs assume that their parameters are changing across time. We design a DLM with model parameters defined at each time unit to describe the spatio-temporal characteristics of time-series traffic data. Based on our DLM and its model parameters analytically trained using historical data, we suggest an optimal linear predictor in the minimum mean square error (MMSE) sense. We compare our prediction accuracy of travel time for freeways in California (I210-E and I5-S) under highly congested traffic conditions with those of other methods: the instantaneous travel time, k-nearest neighbor, support vector regression, and artificial neural network. We show significant improvements in the accuracy, especially for short-term prediction.

Perimeter control schemes proposed to alleviate congestion in large-scale urban networks usually assume perfect knowledge of the accumulation state together with current and future inflow demands, requiring information about the origins and destinations (OD) of drivers. Such assumptions are problematic for practice due to: (i) Measurement noise, (ii) difficulty of measuring OD-based accumulation states and inflow demands. To address these, we propose a nonlinear moving horizon estimation (MHE) scheme for large-scale urban road networks with dynamics described via macroscopic fundamental diagram. Furthermore, we consider various measurement configurations likely to be encountered in practice, such as measurements on regional accumulations and transfer flows without OD information, and provide results of their observability tests. Simulation studies, considering joint operation of the MHE with a model predictive perimeter control scheme, indicate substantial potential towards practical implementation of MFD-based perimeter control.

Local traffic control schemes fall short of achieving coordination with other parts of the urban road network, whereas a centralized controller based on the detailed traffic models would suffer from excessive computational burden. State estimation for detailed traffic models with limited observations and unpredictability of individual driver behavior create additional complications in the applicability of these models for large-scale traffic control. This point toward the need for designing network-level controllers building on aggregated traffic models, which have recently attracted attention through the macroscopic fundamental diagram (MFD) of urban traffic. Under some conditions, the MFD provides a unimodal, low-scatter, and demand-insensitive relationship between vehicle accumulation and travel production inside an urban region. In this paper, we propose MFD-based economic model predictive control (MPC) schemes to improve mobility in heterogeneously congested large-scale urban road networks. For more realistic simulations of urban networks with route guidance actuation-based control, a new model with cyclic behavior prohibition is developed. This paper extends upon earlier works on perimeter control-based MPC schemes with MFD modeling by integrating route guidance type actuation, which distributes flows exiting a region over its neighboring regions. Performance of the proposed schemes is evaluated via simulations of congested scenarios with noise in demand estimation and measurement errors. Results show the possibility of substantial improvements in urban network performance, in terms of network delays and traveled distance, even for low levels of driver compliance to route guidance.

A new gating strategy for concentric cities based on the notion of the macroscopic or network fundamental diagram and the feedback-based gating concept is introduced and successfully tested. Different regions of large-scale urban networks may experience congestion at different levels and times during the peak period. In this paper, the zone, including the initial core of congestion, is considered as the first region, which has to be protected from congestion via gating; eventually, as the congestion continues to expand, the border of an extended network part becomes the second perimeter for gating control. Remarkable extensions while distributing the ordered controller flow to the gated traffic signals in case of low demand or occurrence of spillback are also considered. A greater part of the San Francisco urban network is used as test-bed within a microscopic simulation environment. Significant improvements in terms of network-wide mean speed and average delay per kilometer are obtained compared to the single perimeter gating and non-gating simulation scenarios.

Recent analysis of empirical data from cities showed that a macroscopic fundamental diagram (MFD) of urban traffic provides for homogenous network regions a unimodal low-scatter relationship between network vehicle density and network space-mean flow. In this paper, the optimal perimeter control for two-region urban cities is formulated with the use of MFDs. The controllers operate on the border between the two regions and manipulate the percentages of flows that transfer between the two regions such that the number of trips that reach their destinations is maximized. The optimal perimeter control problem is solved by model predictive control, where the prediction model and the plant (reality) are formulated by MFDs. Examples are presented for different levels of congestion in the regions of the city and the robustness of the controller is tested for different sizes of error in the MFDs and different levels of noise in the traffic demand. Moreover, two methods for smoothing the control sequences are presented. Comparison results show that the performances of the model predictive control are significantly better than a “greedy” feedback control. The results in this paper can be extended to develop efficient hierarchical control strategies for heterogeneously congested cities.

We propose a methodology for identifying queue spillovers in city street networks with signalized intersections using data from conventional surveillance systems, such as counts and occupancy from loop detectors. The key idea of the proposed methodology is that when spillovers from a downstream link block vehicle departures from the upstream signal line, queues discharge at rates smaller than the saturation flow. The application of the methodology on an arterial site and the comparison with field data show that it consistently identifies spillovers in urban networks with signal-controlled intersections. The method is extended to account for the variations in vehicle lengths. We also investigate the significant effect of spillovers in congestion and show that a macroscopic diagram that connects spillovers with vehicle density exists in large-scale congested urban networks.

Following about 40 years of successful deployment of coordinated traffic-responsive ramp control, a new generation is being developed for Minnesota's freeways based on density measurements, rather than flow rates. This was motivated from recent research indicating that the critical value of density at which capacity is observed is less sensitive and more stable than capacity, thereby allowing the opportunity for more effective control. The main goals of the new approach are to delay the onset of the breakdown and accelerate system recovery when ramp metering is disabled due to the violation of maximum allowable ramp waiting times. This is obtained by a dynamic zone partitioning of the freeway network to identify critical bottleneck locations and coordinated balancing of ramp delays, which aims to avoid mainline breakdown. The effectiveness of the new control strategy is assessed by comparison with the currently deployed version of the stratified zone metering algorithm through microscopic simulation of a real 12-mi 17-ramp freeway section. Simulations show a decrease in delays of mainline and ramp traffic and an improvement of 8% in overall system delays while avoiding maximum ramp delay violations.