This paper proposes a sensitivity analysis framework based on set-valued mapping for deep neural networks (DNN) to understand and compute how the solutions (model weights) of DNN respond to perturbations in the training data. As a DNN may not exhibit a unique solution (minima) and the algorithm of solving a DNN may lead to different solutions with minor perturbations to input data, we focus on the sensitivity of the solution set of DNN, instead of studying a single solution. In particular, we are interested in the expansion and contraction of the solution set in response to data perturbations. If the change of solution set can be bounded by the extent of the data perturbation, the model is said to exhibit the Lipschitz-like property. This 'set-to-set' analysis approach provides a deeper understanding of the robustness and reliability of DNNs during training. Our framework incorporates both isolated and non-isolated minima, and critically, does not require the assumption that the Hessian of loss function is non-singular. By developing set-level metrics such as distance between sets, convergence of sets, derivatives of set-valued mapping, and stability across the solution set, we prove that the solution set of the Fully Connected Neural Network holds Lipschitz-like properties. For general neural networks (e.g. Resnet), we introduce a graphical-derivative-based method to estimate the new solution set following data perturbation without retraining.

Traffic simulation is important for transportation researchers, analysts, and policymakers. It can be used to test vehicle/traffic control algorithms, gain insights into traffic dynamics, and develop traffic management strategies that can improve the efficiency and safety of transportation systems. Unfortunately, many existing simulation platforms have limitations to cater to diverse simulation scales. This study presents a comprehensive multiscale vehicle-traffic-demand (VTD) simulation platform tailored for connected and automated transportation systems. This platform integrates Unity 3D, Simulation of Urban Mobility (SUMO), and Multiagent Transport Simulation (MATSim) to facilitate an in-depth analysis of both micro and macro-level traffic behaviors. A critical aspect of our work involves the meticulous setup and calibration of traffic networks in Greater and Downtown Seattle, ensuring effective integration and communication between the various simulation tools. This advanced platform not only serves as a robust tool for testing and refining vehicle/traffic control algorithms but also opens new avenues for research into traffic dynamics learning and the development of sophisticated traffic control solutions.

Fixed-time control is a widely adopted and cost-effective method for signalized intersections. However, existing studies utilizing connected vehicle (CV) data have not effectively addressed fixed-time control due to their reliance on specific vehicle arrival assumptions. To overcome this limitation, this study presents a novel traffic control approach for fixed-time signalized intersections based on a cumulative flow diagram (CFD) framework. The proposed method comprises a CFD model and a multi-objective optimization model. The CFD model establishes analytical relationships between traffic flow operations and varying signal timing parameters, with intersection demand estimated using a novel weighted maximum likelihood estimation method. A multi-objective optimization model based on CFD is formulated to minimize exceeded queue dissipation time as the primary objective and average delay as the secondary objective, which is applicable under both undersaturated and oversaturated traffic conditions. Leveraging the data-driven nature of the CFD model, a specially designed bi-level particle swarm optimization-based algorithm is employed to determine optimal cycle length (and offset if applicable) and green ratios separately. Evaluation results demonstrate that the proposed method outperforms Synchro, a conventional approach, in terms of average delay and queue under various traffic conditions. Moreover, the proposed method exhibits the capability to handle specialized scenarios involving spillbacks.

Accurate and robust localization is crucial for supporting high-level driving automation and safety. Modern localization solutions rely on various sensors, among which GPS has been and will continue to be essential. However, GPS can be vulnerable to malicious attacks and GPS spoofing has been identified as a high threat. With transportation infrastructure becoming increasingly important in supporting emerging vehicle technologies and systems, this study explores the potential of applying infrastructure data for defending against GPS spoofing. We propose an infrastructure-enabled framework using roadside units as an independent, secured data source. A real-time detector, based on the Isolation Forest, is constructed to detect GPS spoofing. Once spoofing is detected, GPS measurements are isolated, and the potentially compromised location estimator is corrected using secure infrastructure data. We test the proposed method using both simulation and real-world data and show its effectiveness in defending against various GPS spoofing attacks, including stealthy attacks that are proposed to fail the production-grade autonomous driving systems.

The recent emergence of new technologies and systems such as connected and automated vehicles (CAVs), novel incentive and routing platforms, and shared mobility services is making a significant impact on traffic flow in road networks. The rapid development of these innovations, powered by new capabilities in data collection, communication, and vehicle autonomy raises both great opportunities and new challenges for managing and controlling the transportation network efficiently. It is thus imperative to integrate the emerging systems into a dynamic transportation network analysis, and to develop new methodologies, which coherently integrate dynamic traffic models with increasingly available data, and methods for large-scale computation. Consequently, they call for new theories, models, computational methods, and application scenarios to study dynamic transportation networks with the emerging technologies as essential components.

Although considerable methods have been developed for the performance evaluation of signalized intersections using sampled vehicle trajectories, most of them aim at estimating a single parameter and cannot describe the entire arrival–departure process of the traffic flow. This significantly constrains the application of these methods for a comprehensive evaluation and efficient optimization of signalized intersections. In this paper, we propose a cumulative flow diagram (CFD) estimation and prediction method using sampled vehicle trajectories. It can be used to calculate multiple performance measures—traffic volume, queue length, average delay, and total delay—based on the estimated CFD for the current signal timing plan. Concurrently, it can be further employed for signal control optimization based on the predicted CFDs for candidate signal timing plans. The core idea of the proposed method is to generate the cumulative arrival curve based on the arrival characteristics of the sampled vehicles, and then fit the queue leaving points to obtain the cumulative departure curve. Thereby, given the current or any candidate signal timing plan, we can estimate or predict the CFDs by updating the sampled vehicle arrivals. The proposed method is evaluated using both simulation and empirical data. The simulation results yield that the average estimation error of the four performance measures is 10.3% under a real-world level penetration rate of 10%. Meanwhile, similar accuracies are achieved for the CFD prediction. The empirical results show that under a penetration rate of 8.6%, the estimation errors of the traffic volume and queue length are 2.7% and 3.3%, respectively.

Customized path-based speed prediction is an eventful tool for congestion avoidance, route optimization and travel time prediction for navigation apps, cab-hailing companies and autonomous vehicles. Traditionally, the speed prediction algorithms are based on road segments and can only support several main roads. Path-based speed prediction is very challenging since the speed is always changing in different path locations and is jointly affected by lots of complicated factors. This article presents a novel deep learning framework for customized path-based speed prediction. A Path-based Speed Prediction Neural Network (PSPNN) is designed to achieve speed predictions for a given path and attributes information. A hierarchical Convolutional Neural Network (CNN) and deep Bidirectional Long Short-Term Memory (Bi-LSTM) structure for different kinds of feature extraction are applied for multiple levels: the path cell, sub-path and the whole path. The method narrows down the prediction unit from road segments to customized path cells (mean length: 59.52m) and achieves a mean absolute error (MAE) of 1.94 m/s and Mean Absolute Percentage Error (MAPE) of 18.14%, showing the potential of serving rigorous data-driven applications. So far, PSPNN is the first made-to-order path-based speed prediction algorithm and can help both travelers and managers to obtain large-scale bespoke paths speed information in advance.

With the rapidly growing deployment of intelligent transportation systems (ITS) and smart traffic applications, vehicle trajectory data are ubiquitously generated, e.g., from GPS navigation systems, mobile applications, and urban traffic cameras. Analyzing such fine-grained data would greatly benefit the development of ITS and smart cities, yet pose severe privacy risks due to the recorded drivers’ visited locations, routes, and driving habits. Recently, some privacy enhancing techniques were proposed to sanitize such data. However, such schemes have some major limitations–they either lack formal privacy notions to quantify and bound the privacy risks, or result in very limited utility, e.g., only a sequence of locations or aggregated information can be released (without retaining the speeds, accelerations and the timestamps of vehicles). In this article, we propose a novel framework to sanitize the fine-grainedvehicle trajectories with differential privacy(VTDP), which provides rigorous privacy protection against adversaries who possess arbitrary background knowledge. Our VTDP technique involves three phases of differentially private sampling, which sequentially generate all the three categories of data (besides a pseudo identity for each vehicle)–position, moving,andtimestamps. It also includes avehicle trajectory interpolationprocedure to further improve the output utility with the properties of fine-grained vehicle trajectory data. We conducted experiments on real vehicle trajectory datasets to validate the performance of our approach.

We study the traffic signal control problem with connected vehicles by assuming a fixed cycle length so that the proposed model can be extended readily for the coordination of multiple signals. The problem can be first formulated as a mixed-integer nonlinear program, by considering the information of individual vehicle's trajectories (i.e., second-by-second vehicle locations and speeds) and their realistic driving/car-following behaviors. The objective function is to minimize the weighted sum of total fuel consumption and travel time. Due to the large dimension of the problem and the complexity of the nonlinear car-following model, solving the nonlinear program directly is challenging. We then reformulate the problem as a dynamic programming model by dividing the timing decisions into stages (one stage for a signal phase) and approximating the fuel consumption and travel time of a stage as functions of the state and decision variables of the stage. We also propose a two-step method to make sure that the obtained optimal solution can lead to the fixed cycle length. Numerical experiments are provided to test the performance of the proposed model using data generated by traffic simulation.

Signalized intersections play an important role in transportation efficiency and vehicle fuel economy in urban areas. This paper proposes a cooperative method of traffic signal control and vehicle speed optimization for connected automated vehicles, which optimizes the traffic signal timing and vehicles' speed trajectories at the same time. The method consists of two levels, i.e., roadside traffic signal optimization and onboard vehicle speed control. The former calculates the optimal traffic signal timing and vehicles' arrival time to minimize the total travel time of all vehicles; the latter optimizes the engine power and brake force to minimize the fuel consumption of individual vehicles. The enumeration method and the pseudospectral method are applied in roadside and onboard optimization, respectively. Simulation studies are conducted to compare the proposed method with benchmark methods. The results show significant improvement of transportation efficiency and fuel economy by the cooperation method.

We study the traffic signal control problem under the connected vehicle (CV) environment by assuming a fixed cycle length so that the proposed model can be extended readily for the coordination of multiple signals. The signal control problem is to minimize the weighted sum of total system fuel consumption and travel times. Due to the large dimension of the problem and the complexity of the nonlinear car-following model, we propose a Dynamic programming (DP) formulation by dividing the timing decisions into stages (one stage for a phase) and approximating the fuel consumption and travel time of a stage as functions of the state and decision variables of the stage. We also propose a two-step method, the end stage cost, and a branch and bound algorithm, to make sure that the obtained optimal solution can lead to the fixed cycle length. Numerical experiments are provided to test the performance of the proposed model using data generated by traffic simulation.

Existing traffic signal optimization and vehicle speed optimization at signalized intersections cannot work together for the lack of proper cooperation methods. We propose the V2I (vehicle to infrastructure) based cooperation between traffic signal and approaching vehicles which optimizes the traffic signal and vehicles' speed trajectories simultaneously. The cooperation consists of roadside traffic signal optimization and onboard speed control, of which the former calculates the optimal traffic signal timing and vehicles' arriving time to minimize trip time and the latter optimizes the vehicle engine power and brake force to minimize the fuel consumption in the whole trip. A simulation study is conducted to compare the proposed cooperation method and the actuated signal control method. The simulation results show significant improvement of transportation efficiency and vehicle fuel economy by using the cooperation method.

Efforts to anonymize collections of location traces have often sought to reduce re-identification risks by dividing longer traces into multiple shorter, unlinkable segments. To ensure unlinkability, these algorithms delete parts from each location trace in areas where multiple traces converge, so that it is difficult to predict the movements of any one subject within this area and identify which follow-on trace segments belongs to the same subject. In this paper, we ask whether it is sufficient to base the definition of unlinkability on movement prediction models or whether the revealed trace segments themselves contain a fingerprint of the data subject that can be used to link segments and ultimately recover private information. To this end, we study a large set of vehicle locations traces collected through the Next Generation Simulation program. We first show that using vehicle moving characteristics related features, it is possible to identify outliers such as trucks or motorcycles from general passenger automobiles. We then show that even in a dataset containing similar passenger automobiles only, it is possible to use outlier driving behaviors to link a fraction of the vehicle trips. These results show that the definition of unlinkability may have to be extended for very precise location traces.

Signal timing information is important in signal operations and signal/arterial performance measurement. Such information, however, may not be available for wide areas. This imposes difficulty, particularly for real-time signal/arterial performance measurement and traffic information provisions that have received much attention recently. We study, in this paper, the possibility of using intersection travel times, i.e., those collected between upstream and downstream locations of an intersection, to estimate signal timing parameters. The method contains three steps: 1) cycle breaking that determines whether a new cycle starts; 2) exact cycle boundary detection that determines when exactly a cycle starts or ends; and 3) effective red (or green) time estimation that estimates the actual duration of the red (or green) time. The proposed method is a combination of traffic flow theory and learning/estimation algorithms and can be used to estimate the cycle-by-cycle signal timing parameters for a specific movement of a signal. The method is tested using data from microscopic simulation, field experiments, and next-generation simulation with promising results.

Traffic monitoring using probe vehicles with GPS receivers promises significant improvements in cost, coverage, and accuracy over dedicated infrastructure systems. Current approaches, however, raise privacy concerns because they require participants to reveal their positions to an external traffic monitoring server. To address this challenge, we describe a system based on virtual trip lines and an associated cloaking technique, followed by another system design in which we relax the privacy requirements to maximize the accuracy of real-time traffic estimation. We introduce virtual trip lines which are geographic markers that indicate where vehicles should provide speed updates. These markers are placed to avoid specific privacy sensitive locations. They also allow aggregating and cloaking several location updates based on trip line identifiers, without knowing the actual geographic locations of these trip lines. Thus, they facilitate the design of a distributed architecture, in which no single entity has a complete knowledge of probe identities and fine-grained location information. We have implemented the system with GPS smartphone clients and conducted a controlled experiment with 100 phone-equipped drivers circling a highway segment, which was later extended into a year-long public deployment.

Automotive traffic monitoring using probe vehicles with Global Positioning System receivers promises significant improvements in cost, coverage, and accuracy. Current approaches, however, raise privacy concerns because they require participants to reveal their positions to an external traffic monitoring server. To address this challenge, we propose a system based on virtual trip lines and an associated cloaking technique. Virtual trip lines are geographic markers that indicate where vehicles should provide location updates. These markers can be placed to avoid particularly privacy sensitive locations. They also allow aggregating and cloaking several location updates based on trip line identifiers, without knowing the actual geographic locations of these trip lines. Thus they facilitate the design of a distributed architecture, where no single entity has a complete knowledge of probe identities and fine-grained location information. We have implemented the system with GPS smartphone clients and conducted a controlled experiment with 20 phone-equipped drivers circling a highway segment. Results show that even with this low number of probe vehicles, travel time estimates can be provided with less than 15% error, and applying the cloaking techniques reduces travel time estimation accuracy by less than 5% compared to a standard periodic sampling approach.