Journal of Intelligent Transportation Systems最新文献

Recognition of spatio-temporal traffic patterns at the network-wide level plays an important role in data-driven intelligent transport systems (ITS) and is a basis for applications such as short-term prediction and scenario-based traffic management. Common practice in the transport literature is to rely on well-known general unsupervised machine-learning methods (e.g., k-means, hierarchical, spectral, DBSCAN) to select the most representative structure and number of day-types based solely on internal evaluation indices. These are easy to calculate but are limited since they only use information in the clustered dataset itself. In addition, the quality of clustering should ideally be demonstrated by external validation criteria, by expert assessment or the performance in its intended application. The main contribution of this paper is to test and compare the common practice of internal validation with external validation criteria represented by the application to short-term prediction, which also serves as a proxy for more general traffic management applications. When compared to external evaluation using short-term prediction, internal evaluation methods have a tendency to underestimate the number of representative day-types needed for the application. Additionally, the paper investigates the impact of using dimensionality reduction. By using just 0.1% of the original dataset dimensions, very similar clustering and prediction performance can be achieved, with up to 20 times lower computational costs, depending on the clustering method. K-means and agglomerative clustering may be the most scalable methods, using up to 60 times fewer computational resources for very similar prediction performance to the p-median clustering.

Dynamic origin–destination (OD) flow is a fundamental input for dynamic network models and simulators. Numerous studies have conducted dynamic OD estimations based on fixed detectors, where a high device coverage rate and data quality are often required to accomplish the desired results. Several existing methods have used probe vehicle trajectories as an additional data source, and generalized least squares (GLS) is commonly recognized as an effective framework. However, the prior matrices used in these models either came from historical data or data obtained by uniform scaling that neglected the variation in penetration rates and suffer from sparsity issues. Moreover, the microscopic information contained in the high-resolution probe vehicle trajectories has not been fully utilized. The possibility of estimating OD flows using only vehicle trajectories without external information is rarely discussed in current literature. Therefore, this paper introduces a dynamic OD flow estimation model solely using probe vehicle trajectories. In the proposed model, two methods based on probe OD pair distribution are proposed to infer prior OD flows. Then the GLS framework is extended by including link travel times as another objective term, and the solution algorithm is adapted to deal with uncertain priors. To validate the proposed model, extensive experiments were conducted on a simulation network. The results show that the proposed model could reliably estimate dynamic OD flows and showed superiority to two existing models. In sensitivity analysis concerning the penetration rate and degree of saturation, the proposed model presented satisfactory performance and could adapt to various conditions.

Timely and accurate detection and recognition of traffic lights are critical for Autonomous Vehicles (AVs) to avoid crashes due to red light running. This paper integrates a new robust machine learning based solution by combining a Convolutional Neural Network (CNN) with computer vision techniques to achieve a real-time traffic light detector. The proposed detection and recognition algorithm is capable of recognizing traffic lights on low-power small-form platforms, which are lightweight, portable, and can be mounted on AVs in daylight scenarios. The LISA open-source dataset is utilized with augmentation methods to increase the accuracy of the solution. The proposed approach achieves 93.42% of accuracy at a speed of 30.01 Frames Per Second (FPS) on an NVIDIA Jetson Xavier platform without using hardware accelerators such as FPGA. This solution is expected to promote the quicker adoption and wider deployment of AVs by increasing the chances of avoiding crashes and ultimately saving lives.

This article explores a novel data-driven approach based on recent developments in Koopman operator theory and dynamic mode decomposition (DMD) for modeling signalized intersections. On signalized intersections, vehicular flow and queue formation have complex nonlinear dynamics, making system identification, modeling, and controller design challenging. We employ a DMD-type approach to transform the original nonlinear dynamics into locally linear infinite-dimensional dynamics. The data-driven approach relies entirely on spatio-temporal snapshots of the traffic data. We investigate several key aspects of the approach and provide insights into the usage of DMD-type algorithms for application in adaptive signalized intersections. To validate the obtained linearized dynamics, we perform prediction of the queue lengths at the intersection and compare the results with the benchmark methods such as ARIMA and long short term memory (LSTM). The case study involves intersection pressure and queue lengths at two Orlando area signalized intersections during the morning and evening peaks. It is observed that DMD-type algorithms are able to capture complex dynamics with a linear approximation to a reasonable extent. The merits include faster computation times and significantly less requirement for a “lookback” (training) window.

To effectively reduce the number of red-light violations and crashes, it is crucial to explore RLR behavior at local intersections, understand the contributing factors, and identify the riskiest intersections by estimating RLR frequency. In this study, a finite mixture modeling method was utilized to understand the contributing factors to RLR behavior and estimate this violating behavior. To develop the RLR estimation models, performance metrics and signal phasing data were collected from the Automated Traffic Signal Performance Measures (ATSPMs) system in two jurisdictions in Arizona: Pima County and the Town of Marana. The results from calibrated models showed that an increase in traffic flow, intersection delay, number of approach lanes, and split failure is associated with an increase in the likelihood of observing red-light violations. In addition, it was found that an increase in cycle length is associated with a decrease in the likelihood of observing the red-light violation. The results of comparing the proposed RLR estimation method with several conventional methods, the Poisson Generalized Linear Model (PGLM), Zero-inflated Poisson Regression Model (ZIPM), and Zero-inflated Negative Binomial Regression Model (ZINB) showed the proposed method outperforms all the models in terms of both model fit and accuracy. The application of the proposed method could be used to analyze the intersections with the highest number of red-light violations. Furthermore, the presented transferability results can be advantageous to transportation agencies within Arizona and urban areas with similar characteristics by providing insight into which model specifications may provide the best RLR estimation accuracy.

Shared Autonomous Vehicles (SAV), or robotaxis, are expected to be commercially available within this decade. This new transport mode has the potential to revolutionize travel, offering a level of service comparable to traditional taxis with much lower prices. This may attract travelers currently using other modes, impacting the economic sustainability of public transport as well as car ownership levels. We investigate this potential demand using a scalable SAV simulation framework. We do not establish a future equilibrium considering the interaction between all users on a detailed road network, but establish the potential demand for a large metropolitan area. Travelers can choose between their current mode and the new SAV mode, with fare and waiting times which depend on real-time demand. For our input data we train a statistical model on a large transport survey from Germany for an urban region, allowing us to generate a large number of trips with realistic characteristics. We conduct a sensitivity analysis to study the effect of several key parameters on the modal shift. We find that SAVs can be attractive to many active mode and public transport users unless regulations are put in place. Our results also show that due to SAV fleet constraints, changes in incentives for travelers currently using cars may have significant consequences on the behavior of other travelers. We further calculate key economic indicators for the fleet, which can inform the discussion on the fleet size and fare level that operators are likely to choose when maximizing their own profit.