Journal of Intelligent Transportation Systems最新文献

In this paper, we propose an estimation algorithm for the shape of the road using trails of leading vehicles via a linear mixed model (LMM) approach. A vehicle trail is essentially the motion trajectory of a vehicle where samples of the historical path are longitudinally collected from the same vehicle at different points in time. Such measurements can be obtained from the fusion system for single or multiple sensor tracking. The aim is to use trails of leading vehicles to depict the road geometry in highway scenarios. The proposed estimation method uses a polynomial-based road model and is built from a LMM, which is one of the most widely used statistical techniques. To avoid the overload of memory usage from trail samples, trail data are first processed by the newly developed compression and chopping mechanisms before being imported into the LMM framework. Moreover, the profile likelihood function is used to alleviate the computational burden and reduce the number of iterations in the Newton-Raphson algorithm in the LMM. Finally, the proposed method is then evaluated by two publicly available next generation simulation (NGSIM) datasets. The large-scale simulation results show that the road shape estimated by the proposed method has the root mean square error (RMSE) less than 0.5 meters in average for all ranges compared with the ground truth road shape. This suggests that our method provides an accurate road shape estimation and captures the shape of the road successfully.

Intelligent traffic control and guidance system is an effective way to solve urban traffic congestion, improve road capacity and guarantee drivers' travel safety, while short-term traffic flow prediction is the core of intelligent traffic control and guidance system. To investigate the long-term memory and the dynamic feature of short-time traffic flow time series, a hybrid model was proposed by integrating autoregressive fractionally integrated moving average (ARFIMA) model and nonlinear autoregressive (NAR) neural network model to predict short-time traffic flow, in which ARFIMA model can address the long-term memory of linear component and NAR neural network can accommodate the dynamic feature of nonlinear residual component. First, the ARFIMA model was employed to predict the linear component of traffic flow, and the results were compared with those of autoregressive integrated moving average (ARIMA) model. Next, the NAR neural network model was adopted to forecast the nonlinear residual components, and the weighted results were considered as the predicted flow of the hybrid model. The proposed hybrid model was validated by using the cross-sectional traffic flow data in California freeways obtained from the open-access PeMS database. The results showed that the ARFIMA model considering the long-term memory can effectively predict the short-term traffic flow, and the prediction accuracy of the hybrid model is better than that of the singular models. The findings provide an alternative for the short-term traffic flow prediction with lower error and higher accuracy.

In this paper, we extend the nonlinear adaptive time gap car-following model to enhance the local and string stability for adaptive cruise control systems considering a time-lag in the lower level vehicle dynamics and a sensor time-delay. Both over-damped local and string stability analyses are performed mathematically and examined by simulation. The over-damped string stability criterion fulfills all the Lp stability norms, where $p\in \left[1,\infty \right].$ Here we consider a time-lag operating in the lower level of the longitudinal control system’s architecture, a sensor time-delay, and heterogeneity in the vehicle dynamics of the platoon. The adaptive time gap model without these attributes is intrinsically stable. However, it turns out that the introduction of a lag, a delay, or heterogeneity in the lower vehicular level reduces the performance in terms of stability, yielding unsafe damped oscillating collective behaviors. Henceforth we extend the model to enhance the stability by transforming the model to a homogeneous structure, without changing the fundamental dynamics. The results show that the extended model satisfies over-damped criteria for both local and string stability, considering actuator time-lag, sensor time-delay, and heterogeneity in the lower level vehicle dynamics. Such features are expected for automated driving systems.

The purpose of this study is to perform analysis, modeling, and simulation (AMS) to investigate the effectiveness of connected vehicle (CV)-based Weather Responsive Management Strategies (WRMS) to address safety concerns on freeway corridors under adverse weather conditions. This study investigates three CV-based WRMS applications: Forward Collision Warning (FCW), Early Lane Change (ELC) advisory, and Variable Speed Limit (VSL), designs operational alternatives for WRMS using CV data, and develops an AMS tool using a weather-sensitive microscopic traffic simulator to understand the effectiveness of the three WRMS under different scenarios. Various CV market penetration rates (MPR), weather conditions, and WRMS algorithm settings are tested in this study. The case study is based on a real-world freeway corridor, a segment of the I–80 Connected Vehicle Testbed in Wyoming. The simulation results show the effectiveness of selected WRMS applications and provide operational insights that state and local transportation agencies may use in future strategic planning and operations of their weather-responsive programs.

The present research aims to understand the safety over the midblock road sections and proposes a safety framework using the conventional Time to Collision (TTC) measure. In the present work, the safety framework underlines a supporting structure connecting the actions of the surrounding vehicles and assesses the collisions changes for a given subject vehicle. The Framework principally checks the likelihood of lateral overlap and the time gap between the subject vehicle and its surrounding vehicles. Later, for the trajectory data development, an automated trajectory data development tool is built with the help of image processing for generating the trajectory data from the study sections. In supporting the developed safety framework, the lateral movement of the vehicles is modeled precisely with the help of deep learning. Further, the conceptualized safety framework is tested with the developed trajectory data sets over the study sections. From the results, it is observed that, in mixed traffic, the collision points are over the entire geometry of the study section. In the case of homogeneous traffic, the collision instincts are clustered toward the median lanes. With the advancement of technology, trajectory data development can be a real-time exercise, and the safety framework can be implemented. By applying the study methodology, the critical spots over the road network can be flagged for better treatment and improve safety over the sections.

Real-world route navigation data indicate that nontrivial portion of drivers do not prefer the system-recommended best routes. Current navigation systems have simplified assumptions about drivers’ route choice preferences and do not adequately accommodate drivers’ heterogeneous route choice preferences, mainly because of: (i) difficulty in acquiring exogenous criteria (e.g., sociodemographic information) that are typically used to differentiate drivers’ preferences in behavioral modeling; and (ii) difficulty in capturing preference of individuals due to limited preference data at the individual level. To address these, this paper introduced a human-centric machine learning technique named Multi-Task Linear Classification Model Adaption (MT-LinAdapt). It can capture drivers’ common aspects of route choice preferences and yet adapts to each driver’s own preference. In addition, any evolvement of individual drivers’ preferences can be simultaneously integrated to update the common preference for further individual drivers’ preference adaptation. This paper evaluated MT-LinAdapt against two state-of-the-art route recommendation strategies including an aggregate-level and an individual-level data-based strategies, which are categorized based on the data used for modeling. With a real-world dataset containing 30,837 drivers’ navigation usage data in Daegu City, South Korea, MT-LinAdapt was compared to existing strategies for its performance at different levels of data availability, and showed at least the same performance with existing strategies when minimum preference data is available and achieves up to 7% higher prediction accuracy as more data becomes available. Higher prediction accuracies are expected to bring better user satisfaction and compliance rates which can further help with transportation system control and management strategies.

Effective performance of arterial corridors is essential to community safety and vitality. Considering the dynamic nature of traffic demand, efficient management of these corridors require frequent updating of the traffic signal timings through various strategies. Agency resources for these activities are commonly scarce and are primarily by public complaints.

This study provides a workflow using probe-based data to measure and compare different segments on arterial corridors in terms of the traffic signal performance measures that can capture travel time dynamics across signalized intersections. The proposed methodology identifies a group of dynamic days followed by evaluation of travel rate based upon remaining non-dynamic days. Dynamic days represent the variability of traffic on a segment. Consequently, a corridor having high number of dynamic segments along with poor performance during normal days would be a candidate for adaptive control. Further, to handle the large-scale data source collected from city-wide or statewide traffic signals, the study adopts parallel computation-based strategy using MapReduce technique. A case study was conducted on 11 corridors within Des Moines, Iowa, to demonstrate the efficacy of the proposed approach, which identified two arterial corridors, Merle Hay Road and University Avenue, to be suitable for adaptive traffic signal control.

Vehicle platooning, a coordinated and controlled vehicle-following strategy, addresses the issue of high fuel consumption of heavy-duty vehicles. This research considers platoons that are formed on the fly in an ad-hoc manner. We investigate two types of ad-hoc platoon formation and corresponding platoon dissolution strategies. The first approach forms a platoon greedily without considering the order of destinations of the platoon members. This approach enables a quick formation but imposes an overhead of platoon rebuilding, and consequently, additional fuel cost when platoon members leave. An alternative approach forms a platoon in the order of the destinations of its platoon members. This ordered approach incurs a comparatively higher formation time due to vehicles’ reorganization but does not lead to further overhead of platoon rebuilding. We investigate whether these ad-hoc formation and dissolution strategies can preserve the original fuel benefit of platooning, and which of the two ad-hoc formation strategies are more fuel-efficient. The experimental results show that the greedy formation of the platoon is more fuel-efficient for a multi-lane highway. The proposed prediction model provides 90.4% prediction accuracy for the greedy approach and 82.2% prediction accuracy for the ordered approach on average, for platoon sizes from two to six vehicles.

Accurate travel time prediction for major freeways and corridors is crucial but challenging when road incidents happen. Data-driven models require a large set of historical data to estimate the spatial and temporal correlations between road incidents and traffic dynamics. More often than not, the amount of historical data under non-recurring conditions is limited when it comes to training the models. This paper investigates the application of data-driven models on an enriched database with simulated travel times. A well-calibrated traffic simulation is used to capture the artificial incident’s impact on a major urban corridor in Sydney, Australia. This procedure is repeated for multiple created incidents, resulting in a synthetic dataset validated by the available actual historical data. Several machine learning models, such as Regression Tree, Support Vector Regression, Extreme Gradient Boosting, and Recurrent Neural Networks are trained and tested based on the simulated travel time and incident information. As a baseline model for comparison, the measured travel time at the prediction time is considered equal to multi-step ahead travel time. Based on the results, the data-driven models developed with the simulated data outperformed the baseline, indicating that our approach can be effectively employed in the travel time prediction.

A novel computational model for the volatile state was developed to improve eco-driving in intelligent transportation systems (ITS). First, the volatile state was divided into eight types using vehicle acceleration and jerk as delineating criteria. Data analysis showed that each jerk type had a different proportion and contribution level to fuel consumption. Next, the model was created by considering eight instantaneous driving decisions as represented by vehicle speed, acceleration, and jerk. The model input included vehicle speed multiplied by acceleration, with jerk as a classifier. The model was calibrated using quadratic polynomial fitting, and validated using another portion of the data. Finally, predictions were compared with the widely used Vehicle Specific Power (VSP) model and the Virginia Tech Microscopic (VT-Micro) model to evaluate model performance. The new model thoroughly captured the measured fuel consumption and provided more accurate predictions in new routes than the above-mentioned models. The mean absolute percentage error value of the new model was ∼4.9% and 3.2% lower than those of the VSP and VT-Micro models, respectively. The determinant coefficient value was up to 95.8%, which was ∼4.6% and 8.5% higher than those of the VSP and VT-Micro models, respectively.