Lagrangian generic second order traffic flow models for node

Macroscopic traffic flow modelling of large-scale freeway networks with field data verification: State-of-the-art review, benchmarking framework, and case studies using METANET

The state of the art of macroscopic vehicle traffic flow models is discussed and the traffic flow stability of these models is analyzed. A nonlinear traffic flow stability criterion is investigated using the wavefront expansion technique. Qualitative relationships between traffic flow stability and model parameters are derived for an entire class of second-order macroscopic traffic flow models. Numerical results axe obtained using the CLAWPACK package for the well-known Payne-Whitham model, in order to illustrate the stability criterion. The newly derived stability results are consistent with previous reported results obtained using both microscopic models and approximate linearization methods. Moreover, the stability criteria derived in the paper can provide more refined information regarding the propagation of traffic flow perturbations and shock waves in second-order models than previously established methodologies.

Connected automated vehicles (CAVs) are typically equipped with communication devices (e.g., dedicated short range communications (DSRC)) and on-board sensors (e.g., Radar, Lidar, Camera, etc.). Communication devices would enable the exchange of real-time information between vehicles and infrastructures via vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) channels. Sensors equipped in vehicles are providing various vehicle sensor data (VSD) such as the CAV’s GPS location, speed and moving direction (trajectory). Existing studies have shown the effectiveness of using CAV trajectories as input in many traffic control models. However, it can be expected that CAVs and human-driven vehicles (HVs) will co-exist on the transportation network in a long period. Hence, to support various traffic control tasks, it is critical to develop a reliable model to understand the real-time traffic pattern in a mixed CAV and HV environment. To satisfy such needs, this project firstly introduces a novel macroscopic traffic flow model which treats CAVs and HVs as separate groups, where a new set of factors are introduced to represent the speed change of HVs due to following CAVs in the traffic stream. Then grounded on the traffic flow model, an optimization function will make real-time adjustment of CAVs’ desired speeds for minimizing the total freeway travel delays. Selecting a segment of I-15 in Salt Lake City as the study site, the research team conducts extensive simulation experiments to evaluate highway performance under different demand levels and CAV penetration rates. Corresponding data analysis reveals that there should exist a critical CAV ratio that can greatly reduce the speed difference between CAVs and HVs in the traffic stream, given the demand pattern. Further discussions have highlighted the need of setting up guidance on highway capacity analysis under the CAV environment.

Positively Conservative Scheme for Macroscopic Traffic Flow Models

We propose a Bayesian approach for parameter uncertainty quantification in macroscopic traffic flow models from cross-sectional data. We consider both a simple first order model consisting in the mass conservation equation and its second order version including a speed evolution equation. A bias term is introduced and modeled as a Gaussian process to account for the traffic flow models limitations. We validate the results comparing the error in the macroscopic variables (flow, speed, density) for both models, showing that second order models globally perform better in reconstructing traffic quantities of interest.

The priority-based traffic light controlling system at an intersection dedicates right-of-way to the highest priority direction. The system is applied for urban traffic system in Vietnam, where various types of vehicles run in close proximity to one another, needs a traffic model suitable to determine the priority of directions going through intersection. The paper has studied current traffic models, microscopic and macroscopic traffic flow models as well as three-phase traffic model, to propose spatio-temporal traffic statistics model. The paper considers red light as a traffic obstacle causing moving jam behind traffic light. The proposed spatio-temporal traffic statistics model studies traffic flow on a road segment bounded by two intersections, where various types of vehicles travel. The spatio-temporal traffic statistics model represents the road surface area occupied by vehicles at locations and over time as spatio-temporal histograms. The expectation of histogram is a factor determining the priority of direction going through intersection.

Real-time travel-time prediction method applying multiple traffic observations

A variety of macroscopic traffic flow models have been applied to formulate the link-based system optimal dynamic traffic assignment (SO-DTA) problem in a many-to-one network. It is expected that SO-DTA models based on various traffic flow models usually result in different optimal link traffic evolution patterns, but whether and how traffic flow models affect the minimal system cost and the corresponding optimal arrival pattern at the destination is unclear. Three traffic flow models–point-queue, spatial-queue, and cell-transmission–are examined, and their resulted minimal system cost in many-to-one networks is compared. It is proved that for three simple networks–a linear network with two sequential bottlenecks, a two-in-one-out merge network, and a one-in-two-out diverge network–the minimal system costs for the SO-DTA models based on these three traffic flow models are identical. Numerical experiments show that this property appears to be held for general many-to-one networks. The reason may be that all three traffic flow models propagate the uncongested flows at the same speed and release the congested flows always at the link capacity. In SO-DTA applications where only the minimal system cost or arrival pattern at the destination is of major interest, the findings can be utilized to substantially expedite the solution procedure.

Macroscopic traffic flow models play a significant role in the Intelligent Transportation Systems (ITS), especially for the urban traffic control and the dynamic traffic assignment. Two main problems limit the application of the existing traffic models. One is the traffic dynamics described by the complex nonlinear equations which are difficult to resolve in real time, and the other is that some model parameters need to be calibrated in prior. This paper presents a hybrid macroscopic and microscopic (macro-microscopic) traffic flow model of freeway traffic by extending the cell transmission model (CTM) with the real time speed dynamics. Simple supplying and demanding functions are defined to represent the traffic propagation between the upstream and downstream cell boundaries. A microscopic car-following model is integrated to specify the dynamics of the average speed in each cell. Simulation results with specific traffic data set collected from a 4-kilometre long urban freeway show the performance of the model compared to the well-known METANET model.

Multi-anticipative bi-directional visual field traffic flow models in the connected vehicle environment

Smartphones and other vehicular sensors equipped with GPS and wireless networking capabilities, are becoming ubiquitous in transportation systems. They provide us with opportunities to gather timely information about road traffic conditions, fuse (assimilate) it with traffic flow models to improve upon the accuracy of these models, and hence supply valuable information for real-time transportation decision making. Macroscopic traffic flow models are described by systems of partial differential equations (PDEs), which are usually only solved numerically. Adaptive moving mesh methods have shown promise in handling high variability of the spatio-temporal features (e.g. shocks and discontinuities) in model's solutions.We propose a novel low-overhead strategy to adaptively select observation sites in real time, by relying on information from the adaptive moving mesh of the numerical solver of the underlying PDEs. The idea is to place more of the limited observational resources to locations of higher variability in the numerical solution. We incorporate our strategy into a particle-filter based data assimilation framework, and compare it with the strategy of gathering and assimilating measurements from evenly spaced observation sites. We experimentally show that our strategy reduces the relative error by up to 53% in estimating vehicle density on a road during phantom jams and traffic jams due to bottlenecks.

In urban road traffic system, a variety of travel modes coexist and interact with each other. Micro modeling and analysis methods are difficult to capture the macro evolution characteristics and interaction of traffic flow. Based on the actual road network and vehicle trajectory data of Beijing, this paper calculates the total number of vehicles on the road section through the taxi trajectory data and the safe distance model to obtain the social vehicle data, and uses the bus moving trajectory data to obtain the bus vehicle data. Based on the three-dimensional macroscopic fundamental diagram (MFD) method, a three- dimensional MFD of social vehicle-bus vehicle is established, and the model parameters are solved by the least square method. The traffic flow data obtained by processing is used to verify its existence. According to the results of regional division, the influence of the number of buses in the two regions on the traffic capacity in the region is compared. The three-dimensional MFD model constructed in this paper can reflect the evolution characteristics of actual traffic flow, which is of great significance for traffic control of different trips.

The main aim of this paper is to use multi-class macroscopic traffic flow and emission models for Model Predictive Control (MPC) for traffic networks. In particular, we use and compare extended versions of multi-class METANET, FASTLANE, multi-class VT-macro, and multi-class VERSIT+. In addition, end-point penalties based on these multi-class traffic flow and emission models are also included in the objective function of MPC to account for the behavior of the traffic system beyond the prediction horizon. A simulation experiment is implemented to evaluate the multi-class models. The results show that the approaches based on multi-class METANET and the extended emission models (multi-class VT-macro or multi-class VERSIT+) can improve the control performance for the total time spent and the total emissions with respect to the non-control case, and they are more capable of dealing with the queue length constraints than the approaches based on FASTLANE. Including end-point penalties can further improve the control performance with a small sacrifice in the computational efficiency for the approaches based on multi-class METANET but not for the approaches based on FASTLANE.

An Automated Calibration Procedure for Macroscopic Traffic Flow Models

In this work, we propose an algorithm performing short-term predictions of the flow and speed of vehicles on a stretch of road, using past measurements of these quantities. This algorithm is based on a physics-aware recurrent neural network. A discretization of a macroscopic traffic flow model (using the so-called Traffic Reaction Model) is embedded in the architecture of the network and yields traffic state estimations and predictions for the flow and speed of vehicles, which are physically-constrained by the macroscopic traffic flow model and based on estimated and predicted space–time dependent traffic parameters. These parameters are themselves obtained using a succession of LSTM recurrent neural networks. The algorithm is tested on raw flow measurements obtained from loop detectors.