Transportation Research Part C-Emerging Technologies最新文献

Short-term origin–destination passenger demand prediction involves modeling spatial and temporal characteristics of urban traffic, such as periodicity in demand rate and directionality in flow path. Meanwhile, spatial and temporal heterogeneities often lead to constantly evolving dynamics in in passenger demand, e.g., passengers may exhibit different mobility patterns at different periods or in different regions. Many models fail to capture these heterogeneities and adjust parameters adaptively, leading to suboptimal prediction results. In this paper, we propose a novel space–time adaptive network (STAN) to address these issues. Spatially, an edge-based backbone with a global receptive field is devised. Edge embeddings directly represent pair-wise relations between regions, preserving more fine-grained information and directional interactions. The backbone adaptively updates edge embeddings by fusing static and dynamic information from origin and destination regions, enabling the model to learn intricate spatial relations from simple input data (i.e., basic relation graphs and historical OD matrices). Temporally, a prompter mechanism is proposed to inject temporal information into model parameters, making them time-dependent. The parameter values exhibit periodicity and continuity for all periods, meanwhile, they can be adjusted for each specific period. It makes the model time-aware and enables it to identify similar periods and differentiate dissimilar ones during training. Extensive experiments are conducted on two real-world datasets (i.e., ten-month taxi trips in New York and one-month ride-hailing trips in Ningbo), and the results demonstrate that our model outperforms baseline models and automatically learns certain spatial and temporal semantics. With its simple yet highly scalable structure, our model proves beneficial for implementations and can assist related tasks such as driver-passenger matching and surge pricing.

Light Detection and Ranging (LiDAR) plays a pivotal role in localization, thereby meeting the imperative to accurately discern vehicle positions and road states for enhanced services in Intelligent Transportation Systems (ITS). As the cooperative perception among multiple LiDARs is necessitated by localization applications spanning extensive road networks, the strategic placement of LiDARs significantly impacts localization outcomes. This research proposes a chance constrained stochastic simulation-based optimization (SO) model for Roadside LiDAR (RSL) placement to maximize the expected value of mean Average Precision (mAP) subject to a budgeted number of RSLs and a chance constraint of ensuring a specific recall value under traffic uncertainties. Importantly, the assessment of a specific RSL placement plan employs a data-driven deep learning approach based on a high-fidelity co-simulator, which is inherently characterized by black-box nature, high computational costs and stochasticity. To address these challenges, a novel Gaussian Process Regression-based Approximate Knowledge Gradient (GPR-AKG) sampling algorithm is designed. In numerical experiments on a bi-directional eight-lane highway, the RSL placement plan optimized by GPR-AKG attains an impressive mAP of 0.829 while ensuring compliance with the chance constraint, and outperforms empirically designed alternatives. The cooperative vehicle detection and tracking under the optimized plan can effectively address false alarms and missed detections caused by heavy vehicle occlusions, and generate highly complete and smooth vehicle trajectories. Meanwhile, the analyses of detection coverage and average effective work duration validate the reasonability of prioritizing the center-mounted RSLs in the optimized plan. The balance analysis of mAP and the number of deployed RSLs confirms the scientific validity of deploying 20 RSLs in the optimized plan. In conclusion, the GPR-AKG algorithm exhibits promise in resolving chance constrained stochastic SO problems marked by black-box evaluations, high computational costs, high dimensions, stochasticity, and diverse decision variable types, offering potential applicability across various engineering domains.

This paper introduces an innovative approach to model the lane-change (LC) process of vehicles by employing parallel learning, seamlessly integrating conventional physical or behavioral models with data-driven counterparts. The LC process is divided into two distinct steps: the LC decision and the LC implementation, each independently modeled. For the LC decision model, a utility-based model is embedded into a neural network. Simultaneously, the LC implementation model incorporates a conventional car-following model, replicating the behavior of the new follower of the lane-changer, within the training process of a long-short-term memory model. Empirical trajectory data collected from unmanned aerial vehicles, which provides detailed information on the vehicles’ lane-changing process, serves as the basis for training and testing the proposed models. Additionally, data from a different site is employed to assess model transferability. Results demonstrate that the proposed models adeptly predict both LC decisions and implementations, outperforming baseline physical and behavioral models, as well as pure data-driven models, in terms of prediction accuracy and transferability. These findings highlight the significant potential of these models in improving the precision of microscopic traffic simulators.

With the growing popularity of transportation network companies (TNC), there remains a gap in network equilibrium models that adequately address the emergence of ride-pooling service that pools two orders into one trip. This challenge arises from the need to enumerate all possible combinations of origin–destination (OD) pooling and sequencing. This paper proposes a network equilibrium framework that integrates ride-hailing platforms’ decision on vehicle dispatching and driver–passenger matching on congested road networks. To facilitate the representation of vehicle and passenger OD flows and ride-pooling options, a layered OD graph is created encompassing ride-sourcing and ride-pooling services over OD nodes. To capture road congestion, a connection between the layered OD graph and road networks is established where the vehicle flows on the OD graph form source demands on road networks. Numerical examples are performed on a toy example and the Sioux Fall network to demonstrate our model and algorithm. The proposed equilibrium framework can efficiently assist policymakers and urban planners to evaluate the impact of TNCs on traffic congestion.

This paper proposes a new methodology for measuring street-level pedestrian density that combines the strengths of image-based observations with the scalability of drive-by sensing. Despite its importance, existing methods for measuring pedestrian activity have several limitations, including high costs, limited coverage, and privacy concerns. To overcome these issues, our approach exploits operation logs generated by dashboard cameras of moving vehicles to estimate pedestrian density for each street, which is validated with data from approximately 3,000 taxis operating in central Tokyo. We produce vibrancy maps for 292 station areas in central Tokyo by leveraging machine learning to estimate pedestrian density in streets where measurement data is scarce. We also evaluate the reliability and coverage of the measurement and illustrate how the measured pedestrian density data can be utilized for assessing the validity of walkability measures. The paper concludes that this approach could provide valuable data to inform urban planning and city operations.

In the last decade, pressure-based schemes such as Back Pressure and Max Weight algorithms have been widely researched and applied for traffic signal control due to their simplicity and proven throughput maximization. In such algorithms, the next chosen signal phase at an intersection in a road network is the one with the highest measured weight, representing the pressure of traffic movements at the intersection, determined based on a single characteristic of the traffic flow or vehicles’ state at that intersection. This paper develops a new optimal Max Weight control mechanism to enhance the network throughput and reduce vehicle delays in a network using a concept of platooning enabled by Connected Vehicles (CVs). To this end, we propose a new proven optimal Max Weight control scheme where the weight consists of several features including the platoon delay, as well as the speed and position of vehicles within the platoon. To the best of our knowledge, this work is the first to propose a platoon pressure-based concept considering multiple configurable attributes in formulating the pressure. Furthermore, we provide a rigorous stability proof that ensures the throughput optimality of the proposed control scheme. In addition, we also develop a machine learning procedure in this paper to optimize the weighting parameter of each attribute contributing to the total pressure enabling its seamless deployment in practice. A number of simulation results demonstrate the feasibility of the learning procedure and show that our Max Weight platoon pressure-based scheme outperforms the state-of-the-art and well-known existing pressure-based algorithms.

A critical challenge that future autonomous driving systems face is improving the ability to cope with complex real-world interaction scenarios such as uncontrolled intersections. In the near future, a mixed traffic flow of human-driven vehicles (HDVs) and connected autonomous vehicles (CAVs) will coexist in transport networks, which motivates us to explore the interaction between HDVs and CAVs to improve traffic efficiency and safety. To help CAVs better interact with HDVs and adapt to the mixed-flow environment, we propose a human-like decentralized control strategy for CAVs. First, a game-theoretic framework is proposed to model multi-vehicle interactions (including HDV-CAV, CAV-CAV interactions) in the mixed-flow environment. The existence of solutions is proven to ensure the feasibility of the proposed game-theoretic model. Next, a driving style recognition algorithm is embedded into the proposed model to help CAVs understand and predict human drivers’ actions. The proposed model is calibrated via a real-world dataset and used to simulate traffic in several testing scenarios. Real-world vehicle trajectories are used to verify the accuracy of generated vehicle trajectories in simulations. Experimental results indicate that 1) CAVs can take more reasonable actions to determine whether to yield while ensuring safety when competing for the right of way with HDVs using the proposed method compared with conservative driving strategies, 2) a higher penetration rate of CAVs can significantly enhance travel efficiency and lower collision risk at uncontrolled intersections.

As many public transportation systems around the world transition to electric buses, the planning and operation of fleets can be improved via tailored decision-support tools. In this work, we study the impact of jointly locating charging facilities, assigning electric buses to trips, and determining when and where to charge the buses. We propose a mixed integer linear program that co-optimizes planning and operational decisions jointly and an iterated local search heuristic to solve large-scale instances. Herein, we use a concurrent scheduler algorithm to generate an initial feasible solution, which serves as a starting point for our iterated local search algorithm. In the sequential case, we first optimize trip assignments and charging locations. Charging schedules are then determined after fixing the optimal decisions from the first level. The joint model, on the other hand, integrates charge scheduling within the local search procedure. The solution quality of the joint and sequential iterated local search models are compared for multiple real-world bus transit networks. Our results demonstrate that joint models can help further improve operating costs by 14.1% and lower total costs by about 4.1% on average compared with sequential models. In addition, energy consumption costs and contracted power capacity costs have been reduced significantly due to our integrated planning approach.

Metro is the main travel model for urban commuters in many metropolises around the world. During peak hours, large numbers of passengers pour into metro stations for rail services, but some are unable to board the trains in time and left stranded on the platform or even queuing outside the stations. The trip reservation (TR) strategy, where passengers preplan their trips and reserve their entry time to the stations. This paper develops an entry reservation strategy (ERS) to optimize the commuter flow during peak hours, and construct a multi-objective passenger flow joint optimization model based on many-to-many passenger demand to minimize the total trip cost of passengers at reservation station and the number of stranded passengers at intermediate stations. The passenger flow optimization problem is formulated as a mixed-integer non-linear programming (MINLP) model. We design an iterative sequential search algorithm combined with the GUROBI solver to obtain the parameters of the optimal ERS and the passenger flow distribution in the metro system after disaggregated reformulation of the complex constraints of the model. We also demonstrate the accuracy and effectiveness of the proposed method with two experiments – an illustrative example and a large-scale case study of Beijing Metro. The results of Beijing Metro experiment show that the joint optimization model with entry reservation strategy (JO-ERS) reduces the number of stranded passengers by 88.46 % compared with the original passenger flow from the AFC.

Better understanding of the equity impacts of automated vehicles (AV) is needed to design equitable deployment plans for AV technology. The goal of this study was to develop and demonstrate a modeling framework to support distributional equity assessments of AV systems prior to their wide adoption. This modeling framework takes a disaggregated approach that integrates agent-based simulation, diverse transportation outcomes, and equity analyses. The framework uses individual-level data to capture detailed disparities in transportation outcomes, applies a state-of-the-art multi-agent traffic simulation model (MATSim) to simulate privately-owned and shared AVs simultaneously, and considers distributional equity from multiple perspectives. To test and demonstrate the framework, a case study was conducted for the Tampa Bay region. Five scenarios were considered with different AV market shares and integration strategies based on scenario planning by the U.S. Federal Highway Administration. Results reveal that high AV penetration rates were required for substantial reductions in inequality. The introduction of AVs led to a more even distribution of accessibility, but slightly more uneven distribution of traditional mobility and affordability. Impacts of disparities in outcomes for disadvantaged groups were mixed. These results suggest that AVs will likely perpetuate existing inequity in the transportation system as long as its fundamental structure remains the same as today. Results highlight the importance of planning and design strategies that directly address the distributional impacts to ensure that AV technology is deployed equitably.