Model-based traffic state estimation using camera-equipped probe vehicles

Aiming at objectively measuring the realism of virtual traffic flows and evaluating the effectiveness of different traffic simulation techniques, this paper introduces a general, dictionary-based learning method to evaluate the fidelity of any traffic trajectory data. First, a traffic pattern dictionary that characterizes common patterns of real-world traffic behavior is built offline from pre-collected ground truth traffic data. The corresponding learning error is set as the benchmark of the dictionary-based traffic representation. With the aid of the constructed dictionary, the realism of input simulated traffic flow data can be evaluated by comparing its dictionary-based reconstruction error with the dictionary error benchmark. This evaluation metric can be robustly applied to any simulated traffic flow data; in other words, it is independent of how the traffic data are generated. We demonstrated the effectiveness and robustness of this metric through many experiments on real-world traffic data and various simulated traffic data, comparisons with the state-of-the-art entropy-based similarity metric for aggregate crowd motions, and perceptual evaluation studies.

Automatically Generating Empirical Speed-flow Traffic Parameters from Archived Sensor Data

The aim of this research was to investigate the performance of simulated traffic data for real-time crash prediction when loop detector stations are distant from the actual crash location. Nearly all contemporary real-time crash prediction models use traffic data from physical detector stations; however, the distance between a crash location and its nearest detector station can vary considerably from site to site, creating inconsistency in detector data retrieval and subsequent crash prediction. Moreover, large distances between crash locations and detector stations imply that traffic data from these stations may not truly reflect crash-prone conditions. Crash and noncrash events were identified for a freeway section on I-94 EB in Wisconsin. The cell transmission model (CTM), a macroscopic simulation model, was applied in this study to instrument segments with virtual detector stations when physical stations were not available near the crash location. Traffic data produced from the virtual stations were used to develop crash prediction models. A comparison revealed that the predictive accuracy of models developed with virtual station data was comparable to those developed with physical station data. The finding demonstrates that simulated traffic data are a viable option for real-time crash prediction given distant detector stations. The proposed approach can be used in the real-time crash detection system or in a connected vehicle environment with different settings.

This paper developed a first-order deterministic macroscopic model H-CTM (heterogeneous cell transmission model) based on basic cell transmission model (CTM). The methodological calibration of the model considers the aggregate behavior of heterogeneous traffic under weak lane discipline. H-CTM incorporates a nonlinear fundamental diagram (FD), modeled for non-lane-based heterogeneous traffic, instead of traditional piecewise linear FD. For validating the model performance, high resolution traffic data were collected from a major arterial roadway (N3-Highway) of Dhaka city covering both peak and off-peak periods. Subsequently, the calibrated FD parameters were incorporated in the newly developed H-CTM to simulate the traffic. The estimated traffic states by the proposed model were then compared with basic model of CTM. Model performances, calculated in terms of mean absolute error based on ground truth values show that the H-CTM can be used to describe the traffic dynamics in the stated traffic system better than the basic model.

Macroscopic models are computationally effective and generate reliable estimates of traffic dynamics based upon the principles of flow conservation and propagation, in which the underlying fundamental diagram (i.e. flow-density relationship) has an important part in capturing the plausible traffic propagation over time and space. This study investigates in practice whether, and in what extent, the calibration of the fundamental diagram affects the performance of macroscopic traffic modelling by using motorway data. For study purposes, a modified version of the first-order Cell Transmission Model (CTM) is adopted in order to allow the use of different fundamental diagrams instead of the traditional triangular one. Traffic data from the MIDAS (Motorway Incident Detection and Automatic Signalling) dataset for M25 motorway in the UK are used for the case studies. The findings of this study demonstrate the significance of selecting the best-suited fundamental diagram to each traffic scenario in order to improve the accuracy of traffic modelling and contribute to the performance analysis and management of motorways.

SummaryDespite its importance in macroscopic traffic flow modeling, comprehensive method for the calibration of fundamental diagram is very limited. Conventional empirical methods adopt a steady state analysis of the aggregate traffic data collected from measurement devices installed on a particular site without considering the traffic dynamics, which renders the simulation may not be adaptive to the variability of data. Nonetheless, determining the fundamental diagram for each detection site is often infeasible. To remedy these, this study presents an automatic calibration method to estimate the parameters of a fundamental diagram through a dynamic approach. Simulated flow from the cell transmission model is compared against the measured flow wherein an optimization merit is conducted to minimize the discrepancy between model‐generated data and real data. The empirical results prove that the proposed automatic calibration algorithm can significantly improve the accuracy of traffic state estimation by adapting to the variability of traffic data when compared with several existing methods under both recurrent and abnormal traffic conditions. Results also highlight the robustness of the proposed algorithm. The automatic calibration algorithm provides a powerful tool for model calibration when freeways are equipped with sparse detectors, new traffic surveillance systems lack of comprehensive traffic data, or the case that lots of detectors lose their effectiveness for aging systems. Furthermore, the proposed method is useful for off‐line model calibration under abnormal traffic conditions, for example, incident scenarios. Copyright © 2015 John Wiley & Sons, Ltd.

Road traffic is important to everybody in the world. People travel and commute everyday. For those who travel by cars (or other types of road vehicles), traffic congestion is a daily experience. One essential goal of traffic researchers is to reduce traffic congestion and to improve the whole traffic system operation and the environment. To achieve this goal, we have to first understand prevailing traffic situations, then perform pro-active traffic control and management. The estimation of traffic states in the past, in the present and in the future plays an important role in traffic management and control systems. This thesis focuses on the development of traffic state estimation approaches, which provide such traffic state information. In road networks, traffic states refer to typical quantities, such as travel times, traffic speeds, traffic flow and density. These quantities reflect the traffic conditions. Based on these data, we are able to find out when a traffic jam starts, or where a traffic accident occurs. However, it is not feasible to get the full picture of traffic states from the current monitoring systems. Due to cost and technical constraints, we can only obtain spatially and temporally discretised traffic data. These traffic data are collected mainly from point-based sensors, such as inductive loops, radars, and cameras. Alternatively, traffic information might be observed by probe vehicles with a selected penetration rate. In all cases, the detected data usually contain errors and noise, which might hinder further analyses. Based on these constraints, this thesis aims to develop a traffic state estimation procedure to solve the foregoing problems and to provide accurate and complete traffic state information. In this procedure, both traffic flow models and the available observation data are used to estimate the most probable traffic states within a data-assimilation framework. Our approach is formulated using a moving observer perspective, resulting in a Lagrangian formulation of traffic state estimation. In the Lagrangian coordinate system, coordinates move with the vehicles. The Lagrangian formulated first-order traffic flow model is applied to describe the evolution of traffic state variables. The proposed Lagrangian formulation of traffic state estimation offers both theoretical and computational advantages over the conventional Eulerian formation. Moreover, this approach can capture the dynamics of multiple vehicle classes by implementing a multi-class traffic flow model. In this thesis, data pre-processing methods are also developed to improve the quality of the observation inputs. Both Eulerian and Lagrangian sensing data are incorporated into the state estimation. The online technique, known as the Extended Kalman Filter (EKF), is applied for data assimilation: this combines traffic model prediction with observation input correction. Importantly, the Lagrangian concept is not restricted to the EKF method with the first-order traffic flow model, but can also be applied to other data-assimilation techniques in combination with more involved macroscopic traffic flow models. A series of experimental studies based on both synthetic and real-world data have been performed to test the proposed methodology. These studies have validated both the mixed-class and the multi-class traffic state estimation methods. The results have demonstrated that the Lagrangian traffic state estimation outperforms the Eulerian approaches in the EKF-based framework, and the multi-class approach further improves the performance of state estimation compared with the mixed-class case. In summary, Lagrangian multi-class state estimation can provide accurate class-specific traffic information for class-specific control applications and traffic management.

The paper is written based on research on 'Dynamic Origin-Destination Matrices From Real-Time Traffic Count Information'. The latest development in automatic traffic count data collection enables us to obtain the traffic count information in a real time or short- time-interval basis. For example, ATCS (Area Traffic Control System) already installed in several large cities in Indonesia, such as: DKI-Jakarta (since 1994), Bandung (since 1997), and Surabaya (since 1998) provided us the real-time or short-time-interval traffic count information for all signalised intersections. This traffic data is updated periodically in a short- time-interval basis (e.g. 5, 15, or 30 minutes time interval). This information is provided at the Traffic Control Centre (TCC) of ATCS project and can be directly and easily accessed at a very low cost through internet. This data is the main input for the short-time-interval Origin- Destination (OD) matrix estimation. Before this type of traffic data is used in the OD matrix estimation process; firstly, these data have to be processed in the Data Processing Interface (DPI). Having it processed; the traffic data will then be ready for estimating the short-time-interval OD matrices. The output of short-time-interval OD matrices together with several practical applications will be the main input for the Real-Time Integrated Traffic Information System (RITIS). This information will be stored in a Website designed specifically and informatively for the purposes of user needs (numerical and graphical). One of the most important information is the best routes from each origin zone to each destination zone which have already considered the effect of congestion. This information will be the main data for the development of the Route Guidance System (RGS) so that each driver can choose his best route through the road network. The best route information will be changed in a short-time-interval basis depending on the traffic condition. This short-time- interval traffic system information will become the public-domain information which can be directly and freely accessed through internet by the users (e.g. Planning Authorities, Traffic Authorities, Department of Public Works, Consultants, Police, drivers, radio stations, and TV stations, other related agencies, etc). Moreover, this approach can also be extended to provide the short-time-interval environmental information.

THE DEVELOPMENT OF THE REAL TIME INTEGRATED TRAFFIC INFORMATION SYSTEM (RITIS) FOR INDONESIA

In this paper, we study the equilibriums of the general Cell transmission models (CTM) and the general Lagged Cell transmission models (LCTM). Theoretical analyses show that different CTMs and LCTMs converge to a two-segment almost piecewise-linear equilibrium state with similar shapes, although the variety of the traffic flow-density relationships (fundamental diagram, FD) and the existence of the lagged effect lead to different patterns of convergence trajectories and speeds. These results indicate the usefulness of the CTMs and LTCMs in practice, especially when we are interested on the formations and dissipations of traffic congestions.

The accurate depiction of the existing traffic state on a road network is essential in reducing congestion and delays at signalized intersections. The existing literature in the optimization of signal timings either utilizes prediction of traffic state from traffic flow models or limited real-time measurements available from sensors. Prediction of traffic state based on historic data cannot represent the dynamics of change in traffic demand or network capacity. Similarly, data obtained from limited point sensors in a network provides estimates which contain errors. A reliable estimate of existing traffic state is, therefore, necessary to obtain signal timings which are based on the existing condition of traffic on the network. This research proposes a framework which utilizes estimates of traffic flows and travel times based on real-time estimated traffic state for obtaining optimal signal timings. The prediction of traffic state from the cell transmission model (CTM) and measurements from traffic sensors are combined in the recursive algorithm of extended Kalman filter (EKF) to obtain a reliable estimate of existing traffic state. The estimate of traffic state obtained from the CTM-EKF model is utilized in the optimization of signal timings using genetic algorithm (GA) in the proposed CTM-EKF-GA framework. The proposed framework is applied to a synthetic signalized intersection and the results are compared with a model-based optimal solution and simulated reality. The optimal delay estimated by CTM-EKF-GA framework is only 0.6% higher than the perfect solution, whereas the delay estimated by CTM-GA model is 12.9% higher than the perfect solution.

Vehicular traffic congestion remains one of the major world-wide sources of productivity and efficiency loss, wasteful energy consumption, and avoidable air pollution. For example it is estimated that in 2007, congestion caused urban Americans to travel an additional 4.2 billion hours and to purchase an extra 2.9 billion gallons of fuel. In this talk I will describe a set of modeling and simulation Tools for Operational Planning (TOPL) developed to provide quick and quantitative assessments of the benefits that Transportation Management Center (TMC) control policies can provide on freeway corridors, in order to decrease congestion. A freeway corridor typically comprises a 40 kilometer freeway segment on a highly populated urban area, together with its adjoining major urban streets or arterials. The movement of vehicles in a corridor is regulated by programmable field control elements including arterial intersection signals, ramp-metering signals, and message signs that announce emergency conditions, set speed limits and tolls, and provide driver information. Traffic data is primarily collected through inductive loop detectors buried roughly every kilometer along the freeways' payment, as well as detectors located in some of the major corridor arterials. TOPL contains a self-calibrated Cell Transmission Model (CTM) traffic macroscopic simulator. This simulator relies on a well-accepted theoretical model of traffic flow; it is parsimonious and does not require parameters that cannot be estimated from traffic data; and has been tested for reliability on several freeways. Moreover, it is fast, running several hundred times faster than real time, which can be used with real-time measurements and statistically predicted short term future traffic demands to keep track of the current freeway traffic state, as well as make short-term predictions. I will also discuss the qualitative behavior of a single freeway based on the CTM, and will focus on several results regarding the structure and stability of the set of equilibrium states in single freeway, including the fact that the freeway decomposes into disjoint contiguous segments demarcated by bottleneck links, with each segment having qualitatively the same behavior. These properties will be further explored in the formulation of traffic responsive and coordinated ramp-metering policies, including a coordinated policy that minimizes travel time, model calibration and missing on-ramp imputation techniques, and congestion and state estimation techniques.

Network-wide traffic state reconstruction: An integrated generative adversarial network framework with structural deep network embedding

Traffic control and guidance is very important to intelligent transportation, which is realized by precise real time traffic state forecasting. CTM (Cell Transmission Model) can describe exactly how traffic shock wave and queue form, yet off-ramp diversion coefficients are set to be time-invariant, which is not conform to the reality that is full of random events. Therefore, CTM could not practically meet the forecast demand. To address this problem, a hybrid modeling strategy based on both CTM and traffic flow diversion coefficients forecast is proposed in this paper. In the proposed method, traffic flow diversion coefficients are firstly forecasted using an advanced NN (Neural Network) with error compensation mechanism, and then CTM is used to forecast traffic state on the basis of the first step's predictions. Simulation results show that, the present model can obtain accurate predictions whose accuracy are 4.6 and 5.8 percent higher than CTM, reaching 92.6 and 93.2 percent, in terms of cell vehicle number and link traffic flow, respectively. For this reason, the present model can practically meet the forecast demand.

To describe the dynamics of traffic flow in the urban link accurately, the waves which generate at intersections are adopted as the influencing factors of traffic flow. Based on the urban traffic waves, a wave-oriented variable cell transmission model (WVCTM) is proposed to illustrate the urban traffic flow. In this model, the average density and length are the state variables. The cells are divided by traffic waves. The upstream cell is the influence area of the waves at the upstream intersection, the downstream cell is the influence area of the waves at the downstream intersection, and the rest is the mediate cell. Consistent with the fundamental diagram and the cell division, the traffic states of urban links are divided into six modes. The variation of modes is explained by hybrid automata. Finally, an experiment is designed to verify the feasibility of WVCTM. The data in the experiment come from the actual scene. Compared with the cell transmission model (CTM) and variable-length CTM (VCTM), WVCTM possesses the valuable performance to predict the traffic states. Likewise, it is rational that WVCTM can correctly illustrate the urban traffic flow.