Henry X. Liu

A calibration procedure for microscopic traffic simulation

Simulation modeling is an increasingly popular and effective tool for analyzing transportation problems that are not amendable to study by other means. For any simulation study, model calibration is a crucial step to obtaining any results from analysis. This paper presents a systematic, multistage calibration and validation procedure for microscopic simulation models. The procedure is demonstrated in a calibration study with a corridor network in the southern California. The model validation results for the study network are also summarized.

Use of Local Linear Regression Model for Short-Term Traffic Forecasting

The traffic-forecasting model, when considered as a system with inputs of historical and current data and outputs of future data, behaves in a nonlinear fashion and varies with time of day. Traffic data are found to change abruptly during the transition times of entering and leaving peak periods. Accurate and real-time models are needed to approximate the nonlinear time-variant functions between system inputs and outputs from a continuous stream of training data. A proposed local linear regression model was applied to short-term traffic prediction. The performance of the model was compared with previous results of nonparametric approaches that are based on local constant regression, such as the k-nearest neighbor and kernel methods, by using 32-day traffic-speed data collected on US-290, in Houston, Texas, at 5-min intervals. It was found that the local linear methods consistently showed better performance than the k-nearest neighbor and kernel smoothing methods.

A general MPCC model and its solution algorithm for continuous network design problem

This paper formulates the continuous network design problem as a mathematical program with complementarity constraints (MPCC), with the upper level a nonlinear programming problem and the lower level a nonlinear complementarity problem. Unlike in most previous studies, the proposed framework is more general, in which both symmetric and asymmetric user equilibria can be captured. By applying the complementarity slackness condition of the lower-level problem, the original bilevel formulation can be converted into a single-level and smooth nonlinear programming problem. In order to solve the problem, a relaxation scheme is applied by progressively restricting the complementarity condition, which has been proven to be a rigorous approach under certain conditions. The model and solution algorithm are tested for well-known network design problems and promising results are shown.

Using Microscopic Simulation to Evaluate Potential Intelligent Transportation System Strategies Under Nonrecurrent Congestion

A microsimulation method is presented for evaluating the effectiveness of potential intelligent transportation system (ITS) strategies under nonrecurrent congestion. The evaluated ITS strategies include incident management, adaptive ramp metering, traveler information systems, arterial management, and a combination of those strategies. These strategies are implemented and evaluated over a road network in Irvine, California, with the microsimulation model PARAMICS. The evaluation results show that all ITS strategies have positive effects on network performance. Because of the network topology (one major freeway with two parallel arterial streets), real-time traveler information has the greatest benefits among all single ITS strategies. However, a combination of ITS strategies can further increase benefits.

Online Recursive Algorithm for Short-Term Traffic Prediction

Short-term traffic prediction is of great importance to real-time traveler information and route guidance systems. Various methodologies have been developed for dynamic traffic prediction. However, many existing parametric studies focus on fixed-size data and presume time-invariant models. A proposed online adaptive model takes into account historical off-line data. A recursive algorithm is used to obtain computational efficiency and reduced storage. The algorithm is extended to a more general and flexible state-space model, and the predictions are computed recursively with a Kalman filter. A maximum likelihood off-line estimate of the noise covariance matrix and transition coefficients matrix is provided, as well as a recursive calculation of the optimal time-variant parameters on line. The result proves that the state-space model with the nonzero noise covariance matrix outperforms the other algorithms with loop detector data on Interstate 405 near Irvine, California.

VISION-BASED STOP SIGN DETECTION AND RECOGNITION SYSTEM FOR INTELLIGENT VEHICLES

The traffic sign detection and recognition system is an essential module of the driver warning and assistance system. A vision-based stop sign detection and recognition system is presented here. This system has two main modules: detection and recognition. In the detection module, the color thresholding in hue, saturation, and value color space is used to segment the image. The features of the traffic sign are investigated and used to detect potential objects. For the recognition module, one neural network is trained to perform the classification and another one is trained to perform the validation. Joint use of classification and validation networks can reduce the rate of false positives. The reliability demonstrated by the proposed algorithm suggests that this system could be a part of an integrated driver warning and assistance system based on computer vision technologies.

Optimal Sensor Locations for Freeway Bottleneck Identification

In the field of traffic operations, accurate performance measures are crucial for many of the intelligent transportation systems applications. Achieving this accuracy and quality requires that network-based roadway sensors are allocated in locations beneficial to traf- fic operations. However, with the budgetary restrictions most transportation agencies face, these roadway sensors cannot be placed as thoroughly as obligatory for ideal accuracy, requiring these agencies to select a limited number of installments that produce the most optimal results. In this article, a nonlinear integer program is proposed to optimally allocate point sensors along a one-directional freeway corridor, given that any pair of adjacent sensors can produce a benefit for bottleneck identification. The objective of this model is to optimize the accuracy of bottleneck identification subject to resource and monetary constraints. This model is nonlinear and, due to a non-constraints. This model is nonlinear and, due to a non- differentiable function, genetic algorithm is applied to find a solution. We demonstrate that on a case study network with bottlenecks at unknown locations, the model successfully allocates sensors in a manner that optimizes bottleneck identification accuracy.

Real-time estimation of arterial travel time under congested conditions

It is well-known that accurate estimation of arterial travel time on signalised arterials is not an easy task because of the periodic disruption on traffic flow by signal lights. It becomes even more difficult when the signal links are congested with long queues because under such situations the queue length cannot be estimated using the traditional cumulative input–output curves. In this article, we extend the virtual probe model previously proposed by the authors to estimate arterial travel time with congested links. Specifically, we introduce a new queue length estimation method that can handle long queues. The queue length defined in this article includes both the standing queue, i.e. the motionless stacked vehicles behind the stop line, and the moving queue, i.e. those vehicles joining the discharging traffic after the last vehicle in the standing queue starts to move. The moving queue concept is important for the virtual probe method because moving queue also influences the manoeuvre behaviour of a virtual probe. We show that, using the ‘event’ data (including both time-stamped signal phase changes and vehicle-detector actuations) collected from traffic signal systems, time-dependent queue length (including both standing queue and moving queue) can be derived by examining the changes in an advance detectors occupancy profile within a cycle. The effectiveness of the improved virtual probe model for estimating arterial travel time under congested conditions is demonstrated through a field study at an 11-intersection corridor along France Avenue in Minneapolis, MN.

Adaptive signal control system with online performance measure for a single intersection

An adaptive signal control system with an online signal performance measure is introduced. Unlike conventional signal control systems, the proposed method uses real-time delay estimation and an online signal timing update algorithm. As a signal performance measure, intersection delay for each phase is measured in real-time via an advanced surveillance system that reidentifies individual vehicles at upstream and downstream stations by using vehicle waveforms obtained from advanced inductive loop detectors. In each cycle, the signal timing plan is optimized based on the delay estimated from the vehicle reidentification technology. The main thrust of the algorithm is the online control capability utilizing direct intersection delay measures. A description of the overall control system architecture and the optimization algorithm is addressed. Performance of the proposed system is evaluated with a high-performance microscopic traffic simulation program, Paramics, and the preliminary results prove the promising properties of the proposed system.

Solving a class of constrained ‘black-box’ inverse variational inequalities

It is well known that a general network economic equilibrium problem can be formulated as a variational inequality (VI) and solving the VI will result in a description of network equilibrium state. In this paper, however, we discuss a class of normative control problem that requires the network equilibrium state to be in a linearly constrained set. We formulate the problem as an inverse variational inequality (IVI) because the variables and the mappings in the IVI are in the opposite positions of a classical VI. In addition, the mappings in IVI usually do not have any explicit forms and only implicit information on the functional value is available through exogenous evaluation or direct observation. For such class of network equilibrium control problem, we present a linearly constrained implicit IVI formulation and a solution method based on proximal point algorithm (PPA) that only needs functional values for given variables in the solution process.