Transportation Research Part C-Emerging Technologies最新文献

Port operational efficiency and energy consumption are pivotal, but sometimes contradictory factors influencing its competitiveness. In light of this, the simultaneous optimization of these two objectives within the port integrated scheduling of quay cranes, internal vehicles, and yard cranes, can aid in sustaining port development in the era of digitalization and autonomy. Furthermore, given the persistent fluctuations in uncertain operation time of the cranes and vehicles in port, it becomes imperative to consider the robustness of their scheduling plans collectively. This paper therefore aims to develop a new tri-objective mixed-integer programming model for the first time that enables the incorporation of operational uncertainty and energy efficiency into the context of port operation scheduling consideration. The three objectives are makespan, energy consumption, and scheduling plan robustness, which is represented by anti-cascade and robustness evaluation indices. To effectively address complex optimization challenges, a novel multi-objective solution algorithm has been developed, featured with a dynamic fitness evaluation method selection mechanism. This mechanism utilizes a new crowding distance operator based on the cosine distance of objective value vectors to enhance population diversity in the early stages of the algorithm’s iterations. At the later stages, it employs a fuzzy correlation entropy operator to ensure rapid convergence and high-quality solutions. Comparative experiments conducted in scenarios involving emerging technologies such as U-shaped ports and double-cycling operational mode demonstrate the evident improvements achieved by the new model in terms of makespan, energy consumption, and computational efficiency. Based on the compelling experimental results, meaningful insights and implications are put forward, including the potential time and energy savings in port operations, and the practical applicability of these models and algorithms in both port and various other industries.

Accurate prediction of ship emissions aids to ensure maritime sustainability but encounters challenges, such as the absence of high-precision and high-resolution databases, complex nonlinear relationships, and vulnerability to emergency events. This study addresses these issues by developing novel solutions: a novel Spatiotemporal Trajectory Search Algorithm (STSA) based on Automatic Identification System (AIS) data; a rolling structure-based Seasonal-Trend decomposition based on the Loess technique (STL); a modular deep learning model based on Structured Components, stacked-Long short-term memory, Convolutional neural networks and Comprehensive forecasting module (SCLCC). Based on these solutions, a case study using pre and post-COVID-19 AIS data demonstrates model reliability and the pandemic’s impact on ship emissions. Numerical experiments reveal that the STSA algorithm significantly outperforms the conventional identification standard in terms of accuracy of ship navigation state identification; the SCLCC model exhibits greater resistance against emergency events and excels in comprehensively capturing global information, thus yielding higher accurate prediction results. This study sheds light on the changing dynamics of maritime transport and its impacts on carbon emissions.

Understanding the causal impact of meteorological variations on traffic conditions (e.g., traffic flow and speed) is crucial for effective traffic prediction and management, as well as the mitigation of adverse weather effects on traffic. However, many existing studies focused on establishing associations between meteorological situations and traffic, rather than delving into causal relationships, especially with deep learning techniques. Consequently, the ability to identify specific meteorological conditions that significantly contribute to traffic congestion or delays is still limited. To address this issue, this study proposes the Meteorological-Traffic Causal Inference Variational Auto-Encoder Model (MT-CIVAE) to estimate the causal impact of fine-grained meteorological variations (e.g., rain and temperature) on traffic. Specifically, MT-CIVAE is based on the Variational Auto-Encoder and consists of an encoder to recover the distribution of latent confounders and a decoder to estimate the conditional probabilities of treatments. Transformer encoder layers are incorporated to analyze the spatial and temporal correlations of historical traffic data to further enhance the inference capability. To evaluate the effectiveness of the proposed approach for causal inference, real-world traffic flow and speed datasets collected from California, along with corresponding fine-grained meteorological datasets, are employed. The counterfactual analysis is conducted using artificially generated meteorological conditions as treatments, which allows for the simulation of hypothetical meteorological scenarios and the evaluation of their potential impact on traffic conditions. This study develops deep learning methods for assessing the causal impact of meteorological variations on traffic dynamics, offering explanations and insights that can assist transportation institutions in guiding post-meteorology traffic management strategies.

Tracking the source of congestion, namely where the congested traffic flow comes from and goes to, is a key prerequisite to understanding the causes of traffic congestion and facilitates more efficient strategies. In this paper, we track the congestion source by estimating the path flow passing through the congested link. A probabilistic sensor flow assignment model is first developed to infer the whereabouts of each vehicle converging into the congestion. Unlike classical path flow estimation methods, we view path flow as the assigned results of sensor flows rather than OD flows. With this new perspective, an assigned rule, which incorporates route choice preference of drivers and spatial–temporal constraint of vehicular trajectory, is constructed to output more realistic assignments. Moreover, as this model finds most possible destination-path combinations rather than partial paths as assigned results, the complete trip of tracking vehicles, including both driving paths and ODs, can be reconstructed. With the reconstructed trips, disaggregated and hybrid path flow estimation methods are developed to track the source of traffic congestion on the bottleneck link.

The open-source pNEUMA dataset is employed to test the proposed and benchmark methods. It demonstrates that our methods can produce a more realistic traffic pattern for congestion tracking. Significant improvements in estimation accuracy have been achieved with the use of sensor flow assignment model. The proposed disaggregated method has also been tested with a city-scale road network. Experiment results demonstrate that our method is more robust to the uncertainty caused by possible destinations than benchmark.

Emerging technologies drive the ongoing transformation of Intelligent Transportation Systems (ITS). This transformation has given rise to cybersecurity concerns, among which data poisoning attack emerges as a new threat as ITS increasingly relies on data. In data poisoning attacks, attackers inject malicious perturbations into datasets, potentially leading to inaccurate results in offline learning and real-time decision-making processes. This paper concentrates on data poisoning attack models against ITS. We identify the main ITS data sources vulnerable to poisoning attacks and application scenarios that enable staging such attacks. A general framework is developed following rigorous study process from cybersecurity but also considering specific ITS application needs. Data poisoning attacks against ITS are reviewed and categorized following the framework. We then discuss the current limitations of these attack models and the future research directions. Our work can serve as a guideline to better understand the threat of data poisoning attacks against ITS applications, while also giving a perspective on the future development of trustworthy ITS.

Connected automated vehicles (CAVs) have untapped potential to regulate mixed traffic by orchestrating the movement of human-driven vehicles (HVs) at intersections. This research introduces a new controller role for CAVs as regulators of mixed traffic in connected environments. Coordinating the movement of HVs by synchronizing the speed and alignment of CAVs acting as platoon leaders at intersections is a stochastic process with state transition probabilities that vary with traffic speed and vehicular density at the network level. We tackle the problem of regulating mixed traffic at intersections at a macroscopic scale and develop a stochastic model to enhance the operation of mixed traffic consisting of HVs and CAVs by optimizing traffic speed and vehicular density at the network level. Traffic speed and vehicular density are interdependent and vary together at the network level. Therefore, we employ the concept of the network Macroscopic Fundamental Diagram (MFD) to optimize vehicular density by adjusting the spacing between vehicle platoons, led by CAVs, to maximize intersection capacity and network flow at a larger scale. The proposed model is premised on a first-in-first-out reservation-based approach developed for coordinating the movement of vehicle platoons across multiple lanes moving together in cohorts, led by CAVs, at intersections. We account for the randomness in the size, alignment, and arrival time of platoons at intersections in heterogeneous traffic conditions and develop a Markovian approach to capture the stochasticity in modeling the coordination process at intersections. We capture the interrelationship between traffic speed, vehicular density, and inter-cohort spacing at the network level and estimate the upper bound of the flow as a function of density under different CAV penetration rate scenarios. Our numerical results show that optimizing traffic speed and density by adjusting the average spacing between platoons led by CAVs, when the CAV penetration rate in mixed traffic is as low as 20%, can increase the network flow up to 54% of the maximum capacity achievable under uniform CAV traffic conditions.

Attention, or more generally, driver monitoring systems have been identified as a necessity to address overreliance on driving automation. However, research suggests that monitoring systems may not be sufficient to support safe use of advanced driver assistance systems (ADAS), also evidenced by a recent major recall of Tesla’s monitoring software. The objective of the current study was to investigate whether different training approaches improve driver behaviour while using ADAS with an attention monitoring system. A driving simulator study was conducted with three between-subject groups: no training, limitation-focused training (highlighted situations where ADAS would not work), and responsibility-focused training (highlighted the driver’s role/responsibility while using ADAS). All participants (N = 47) experienced eight events which required the ego-vehicle to slow down to avoid a collision. Anticipatory cues in the environment indicated the potential for the upcoming events. Event type (covered in training vs. not covered) and event criticality (action-necessary vs. action-not-necessary) were within-subject factors. The responsibility-focused group made fewer long glances (≥ 3 s) to a secondary task than the no training and limitation-focused groups when there were no anticipatory cues. Responsibility-focused training and no training were associated with faster takeover time at the events than limitation-focused training. There were additional benefits of responsibility-focused training for events that were covered in training (e.g., higher percent of time looking at the anticipatory cues). Overall, our results suggest that even if attention monitoring systems are implemented, there may be benefits to driver ADAS training. Responsibility-focused training may be preferable to limitation-focused training, especially for situations where minimizing training length is advantageous.

While data plays an important role in transportation research, sampled data is not always reliable. Data reliability issue is significant especially for minority groups. In this study, a districting approach is proposed which improves data reliability through aggregation of basic spatial units (BSU), adapted from a max-p-regions problem. The model generates as many aggregated zones as possible that minimize intrazonal heterogeneity while minimizing data margin of error (MOE) of all aggregated zones using a controlling MOE threshold. The problem is first formulated as an integer programming which selects optimal set of zones from a pre-generated set of candidate zones. The difficulty of solving the formulation lies in the generation of the candidate set, so a heuristic solution algorithm is proposed. Two case studies are provided to illustrate the method and validate its performance by evaluating the resulting data quality in an example subsequent planning model. First is an area in Downtown Manhattan with 62 census tracts, comparing the aggregated zones with Neighborhood Tabulation Areas (NTAs) and Taxi Zones. Second is the generation of the New York City Equitable Zoning (NYCEZ), which generated 574 Equitable Zones that reduce the average MOE% of demographic data by 48% for seniors, 75% for low-income population, and 46% for long commuters, all with a district number that is higher than NTAs (2 2 1) and Taxi Zones (2 6 3). NYCEZ and census tracts are then compared in a subsequent model, synthetic population generation, showing an improvement of 6.2% in standard deviation across simulated populations under the proposed zone design. NYCEZ showed smaller variation in the generated population data. The algorithm can help the decision making of public agencies and the service design of mobility providers by producing reliable and equitable data. The algorithm can also be applied to data-sharing between mobility providers and agencies to alleviate privacy concerns.

Predicting the future trajectories of traffic agents is a pivotal aspect in achieving collision-free driving for autonomous vehicles. Although the overall accuracy of existing prediction methods appears promising, most of them overlook the long-tailed challenge in trajectory prediction. They tend to excuse or overlook the disastrous performance in rare yet safety-critical tail events. This paper puts forward a novel framework called hierarchical wave-semantic contrastive learning (Hi-SCL), which attempts to fight the long-tailed challenge in the trajectory prediction task. Our approach innovatively represents each traffic scene as “waves”, and implicitly models traffic multi-stream interactions through wave superposition at both local and global levels. This pioneering incorporation of the wave concept enhances the in-depth comprehension of the traffic scene. On this basis, we introduce the feature hierarchical reshaping method, empowering our network to cope with formidable infrequent cases effectively. This module maintains a collection of feature-enhanced hierarchical prototypes, dynamically steering trajectory samples closer or pushing them farther away in an unsupervised learning setup. Extensive experiments on real-world datasets validate Hi-SCL’s robust overall prediction performance and its effectiveness in addressing long-tailed challenges. Compared to several baseline models, Hi-SCL demonstrates remarkable improvements in general predictive accuracy, with long-term prediction error reductions ranging from 14% to 54% for minADE and 27% to 79% for minFDE. The outcomes of long-tailed experiments further underscore the capacity of Hi-SCL, offering accuracy gains ranging from 2% to 17% in tailed samples. The thorough empirical analyses confirm Hi-SCL’s exceptional capability of wave-semantic representation learning and its effectiveness in reshaping the feature space via hierarchical contrastive learning mechanisms. The proposed new paradigm paves the way for substantial advancements in trajectory prediction, especially in overcoming long-tailed issues, bringing us closer to realizing safer autonomous driving systems.

To model the car following (CF) behavior of mixed traffic flow composed of autonomous vehicles (AVs) and human-driving vehicles (HDVs) in the future, this paper calibrated the lower control algorithm of AVs and proposed a model named theory-data dual driven stochastically generative adversarial networks (TDS-GAN) to describe the CF behavior of HDVs based on experimental data from mixed traffic flow. Firstly, the experimental scenario and collected data were introduced. Secondly, two transfer functions for the lower control algorithm of AVs were compared. Then, to account for the stochasticity of HDVs, the idea of physics-informed deep learning (PIDL) was used to improve generative adversarial networks (GAN) and integrate it with two-dimensional intelligent driver model (2D-IDM). Finally, the effectiveness of the model was verified from both a micro prediction and a macro simulation perspective. The influence of AVs on the stability of mixed traffic flow at different Market Penetration Rate (MPR) was also observed. The results show that TDS-GAN can effectively describe the following behavior and stochasticity of HDVs. When combined with CF model of AVs, it can depict the evolution of mixed traffic flow more accurately. Additionally, AVs can improve traffic flow stability under different penetration rates, and the effect is more significant with higher MPR. However, the introduction of AVs may not necessarily be positive in terms of road capacity.