Zi-You Gao’s research while affiliated with Beijing Jiaotong University and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (132)


Energy-Efficient Train Timetable Under Operational Uncertainty: A Two-Stage Stochastic Model with Customized Benders Decomposition
  • Preprint

January 2024

·

42 Reads

Deheng Lian

·

Zebin Chen

·

Pengli Mo

·

[...]

·








Graph Representation Learning in the ITS: Car-Following Informed Spatiotemporal Network for Vehicle Trajectory Predictions

January 2024

·

6 Reads

·

8 Citations

IEEE Transactions on Intelligent Vehicles

Multimodal synchronization has become the research highlight of the ITS, where complex driving scenarios, various types of vehicles and diverse data sources are crucial constituents. As real-time microscopic traffic characteristics can be vividly represented by graph data, we strive to achieve accurate trajectory predictions via graph-structured series for the stability and the efficiency of transportation. Although the existing data-driven algorithms have achieved fabulous accuracy in various simulation tasks, there are limitations in the distribution and the number of vehicles under investigation. With car-following patterns applied as physical information, we derive the adjacency matrix and design graph filters to explore the spatial dependence between vehicles via the graph-represented multi-lane traffic. The multi-head attention layer is attached to the spatiotemporal convolutional network as an extension. The rationality and the superiority of our model are validated on two calibrated datasets. Through error comparisons, we discuss the role of changeable hyper-parameters to deduce the optimal model for one-step and multi-step predictions. Novel ideas are shared in this paper to simplify the complexity of trajectory prediction in the synchronized transportation system.


On a Flexible Car Use Restriction Policy: Theory and Experiment

March 2023

·

153 Reads

·

2 Citations

Transportation Science

Car use restrictions have been adopted in some mega cities that experience rapid car ownership increase and worsening traffic congestion. Although easy to implement and considered fair, most implementations of this travel demand management policy do not offer travelers the flexibility to choose the days that they cannot use their cars. In this paper, we study a flexible car use restriction policy under which a private car cannot be driven on a certain day of a week, but the day can be chosen by its owner. Under this flexible policy, individuals face a dilemma between driving in congestion and traveling without a car, each incurring a cost of its own. The resulting equilibrium solutions under these two competing choices were derived, and a series of laboratory experiments were carried out to validate the theoretical results. The experimental results are found to be in agreement with the theoretical results. Moreover, our analysis shows that the flexible car use restriction policy reduces the average travel cost with a lesser increase in average driving cost when compared with the traditional car use restriction policy. Funding: Z.-Y. Gao was supported by the National Natural Science Foundation of China [Grant 71621001]. W.-X. Wang was supported by the National Natural Science Foundation of China [Grant 71631002]. R. Jiang was supported by the National Natural Science Foundation of China [Grant 71931002]. X.-Y. Sun was supported by the National Natural Science Foundation of China [Grant 71961002]. B.-Y. Zhang was supported by the National Natural Science Foundation of China [Grants 71922004 and 72131003]. X. Han was supported by the National Natural Science Foundation of China [Grant 71801011] and by the China Postdoctoral Science Foundation [Grant 2018M631331]. Supplemental Material: The e-companion is available at https://doi.org/10.1287/trsc.2023.1200 .



Citations (65)


... Benefiting from the potent deep learning, pioneering work in vehicle trajectory prediction has addressed some of the above challenges. Variational Autoencoders (VAE) [4], Generative Adversarial Networks (GAN) [5], and Graph Neural Networks (GNN) [6] have been utilized to learn trajectory representations and generate multiple possible trajectory samples, effectively capturing multimodal features. These techniques model the complex relationships between vehicles and capture social interactions, leading to more accurate trajectory predictions. ...

Reference:

STS-GAN: Spatial-Temporal Attention Guided Social GAN for Vehicle Trajectory Prediction
Graph Representation Learning in the ITS: Car-Following Informed Spatiotemporal Network for Vehicle Trajectory Predictions
  • Citing Article
  • January 2024

IEEE Transactions on Intelligent Vehicles

... Positive Improve urban transportation conditions [76][77][78] Willing to accept information prompts [79][80][81] Provide travel decision-making assistance [35,[82][83][84][85] Willing to face risk with the help of information [86] Reduce emissions, protect environment [83] Information provided by mobile devices affects most [87] Negative Bad effects of improper information dissemination [88] Unwilling to accept information prompts [89][90][91][92][93][94] Real road network information may not lead to best traffic distribution [95] Information prompts do not lead to traffic condition improvement [96] A majority of travelers are indifferent to information prompts [97][98][99] Better to follow intuition than follow information prompts [100] Neutral Interaction in social platforms is also a part of travel information [101,102] Different attitudes towards information prompts [103][104][105][106] Information prompts may not always have a fixed effect [107,108] Effectiveness varies depending on the penetration rate of ATIS [109,110] Information dissemination has different strategies [111] Differentiated information dissemination considering different personalities of receivers [112] Correct and incorrect information may both have good effects [113] Different impact of information prompts inside/outside congestion area [114] 3. ...

Reducing strategic uncertainty in transportation networks by personalized routing advice: A route-choice laboratory experiment
  • Citing Article
  • January 2024

Travel Behaviour and Society

... Nem pusztán az lesz a változás, hogy az autóinkból eltűnhet a kormánykerék: az önvezető járművek rendszerbe szervezve teljesen átalakíthatják a jelenleg ismert városi mobilitást és városképet (Zuti -Lukovics 2022). Mindezt úgy, hogy egy ideig vegyesen lesznek jelen a hagyományos és önvezető járművek a forgalomban (Xing et al. 2022, Brovarone et al. 2021, és ezen átmeneti időszak hossza, csakúgy, mint a piaci penetrációs szint jelentősen eltérő egyes kutatások eredményei szerint (Milakis et al. 2017, Litman 2023, ETRAC 2019, Bazilinskyy et al. 2019. Például az összekapcsolt autonóm járművek (connected autonmous vehicles, CAV) az útfelületek hatékonyabb kihasználása által lehetőséget biztosít arra, hogy utcáinkat úgy alakítsuk át, hogy azok jobban járhatóak legyenek, és hogy változatosabb tevékenységeknek adjanak helyet, amelyek az utcákat élettel teli városi terekké alakítják át (NACTO 2019, Riggs et al. 2020). ...

Do bicyclists and pedestrians support their city as an autonomous vehicle proving ground? Evidence from Pittsburgh
  • Citing Article
  • October 2022

Case Studies on Transport Policy

... These models are important for the automatic cruise control systems of vehicles and are considered key foundations for intelligent transportation system strategies [4,5]. For half a century, to reflect the actual traffic phenomenon accurately and effectively, researchers worldwide have proposed many mathematical models for car-following analysis [6,7]. These models are typically divided into macro models (such as the hydrodynamic model) [8], the mesoscopic models (such as the lattice model) [9], and the microscopic models (such as the car-following model) [10]. ...

Stochastic factors and string stability of traffic flow: Analytical investigation and numerical study based on car-following models
  • Citing Article
  • November 2022

Transportation Research Part B Methodological

... Steadiness: In real-world applications, the stability of the traffic signal plans is crucial as it ensures that the signal plan does not change dramatically over a short period. We utilize the second-order difference to quantify stability, as it effectively captures the fluctuation in a time sequence's alterations [53]. This metric not only measures the variability of signal changes but also serves as an index of overall system stability, where fewer changes denote higher stability. ...

Stability analysis of stochastic second-order macroscopic continuum models and numerical simulations
  • Citing Article
  • October 2022

Transportation Research Part B Methodological

... The corresponding traffic microsimulation scenarios become even more complex providing often, questionable outputs. New microsimulation models are regularly proposed in the literature for modeling the longitudinal movement of Adaptive Cruise Control (ACC) systems, Cooperative-ACC (CACC) systems, entirely autonomous vehicles or mixed traffic systems with human-and ACC-driven vehicles (Flores and Milanés, 2018;Milanés et al., 2014;Shang et al., 2022;Zheng et al., 2022a). Moreover, to the authors' best knowledge, only few studies in the literature provide insights on the calibration framework of ACC carfollowing models with real-world data, and mostly without providing the insights on the behavioral pattern of ACC systems in the calibration. ...

Oscillation Growth in Mixed Traffic of Human Driven and Automated Vehicles in Both Experimental Study and Simulation
  • Citing Article
  • January 2022

SSRN Electronic Journal

... However, once the BTB IR reached a sufficiently high level, the bus-to-bus interchange began to complement the metro-to-bus interchange again, indicating a synergistic effect. This finding highlights a potential mode competition between bus-to-bus and bus-to-metro interchanges during weekdays, necessitating the optimization of bus lines to enhance cooperation and reduce redundancy (Yang et al., 2022b). The competition could be further complicated by fare differences, as passengers are sensitive to cost variations between bus and metro services, which can significantly influence their mode choice (Gadepalli et al., 2022). ...

Competition and coordination in public transport: A mode choice experiment
  • Citing Article
  • October 2022

Transportation Research Part C Emerging Technologies

... The study found that platform integration can lead to greater social welfare but may not necessarily yield higher profits for platforms. [26] extended the third-party integrator concept, where the matching between passengers and platforms depends on both expected and actual waiting times, demonstrating that the third-party integrator can reduce passengers' trip costs and attract more passengers under certain conditions. Note that all aforementioned works focus on the transportation of people. ...

Order Assignment in a Ride-Sourcing Market with a Third-Party Integrator
  • Citing Article
  • January 2022

SSRN Electronic Journal

... The paper [33] developed an optimization multi-module model by utilizing the accessible route in the primary train schedule. This model solves a repeatable framework of the Lagrangian relaxation by the branch-and-bound algorithm. ...

Joint Optimization of Train Scheduling and Routing in a Coupled Multi-Resolution Space-Time Railway Network
  • Citing Article
  • January 2021

SSRN Electronic Journal

... The second is to model 'integrated' travel choices or a 'bundle' of travel decisions, as travel choice behavior is complex process encompassing multiple dimensions such as whether to travel, destination choice, mode choice, departure time choice, and route choice. Examples of potential scenarios include the joint decision-making of departure time and travel mode in bottleneck problems (Han et al., 2021) and the combined decision-making of travel mode and route choice in the traffic mixing human-and machine-driven vehicles (Wang et al., 2020), etc. Moreover, the experiment-based findings also inspire theoretical modeling at both the micro-individual and macro-network levels, better linking theoretical travel-choice models with real-world travelers' DTD route choice behaviors. ...

The value of pre-trip information on departure time and route choice in the morning commute under stochastic traffic conditions
  • Citing Article
  • October 2021

Transportation Research Part B Methodological