BRBin Ran

Bin Ran

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Bin Ran has a total of 29 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    How Corner Is a Corner Case? Percentile Control for Highway Scenario Generation

    by Jiaxi Liu, Hang Zhou, Hangyu Li, Yifan Wang, Keke Long, Chengyuan Ma, Bin Ran, Xiaopeng Li

    Generating corner-case scenarios with appropriate adversity in a simulation environment is critical for testing an autonomous vehicle (AV) software stack’s safety performance before deployment. Existing autonomous-driving scenario generators can enforce specific behavior, adversity, or feasibility conditions, but they provide limited control over how extreme a generated scenario is relative to plausible futures in the same traffic context. This study represents the adversity of a generated scenario as its percentile in the conditional distribution of future risk given the observed history. This view supports calibrated answers to two questions: how ”corner” a generated corner-case scenario is and how its ”cornerness” can be fine-tuned.To this end, we formulate history-conditioned risk-percentile requests and learn a reference risk distribution that maps each requested percentile to a physical risk target. We then use a percentile-conditioned joint diffusion model with sampling-time risk guidance to generate multi-agent futures, together with a reference-based criterion for evaluating percentile realization.Experiments use the minimum post-encroachment time (PET) between the ego and its surrounding vehicles as the risk surrogate on highD. On the primary evaluation set, our method realizes 1,422 of 1,440 requests within a 0.05 percentile tolerance (98.75%), with mean percentile error 0.00673 and PET-target error 0.00991 seconds. The resulting interface connects context-relative risk specification, physical realization, and evaluation through a common risk scale. Project website and videos of generated scenarios are available at https://hhjj233.github.io/CornerPercentile/.

  • Open Access

    CrossTraffic: An Open-Source Framework for Reproducible Transportation Analysis and Executable Knowledge Management

    by Rei Tamaru, Bin Ran

    Transportation design is governed by codified standards whose knowledge remains locked in documents and closed software. Large language models offer natural-language access to this knowledge but generate their answers, so methods and arithmetic are predictions that cannot be verified against the governing procedures. This paper presents CrossTraffic, a framework that represents regulatory knowledge as an executable knowledge graph. Each provision is implemented as a citation-bearing function and linked to others by its data dependencies, so the standard itself computes the quantities it constrains and traces every result to its source clause. This carries knowledge management for engineering standards from representation and retrieval to execution. Three reasoning operations run over the substrate. Design repair searches the dependency graph for a minimal compliant change and reports infeasibility when none exists. Inverse design synthesizes geometries that meet a target level of service. Defeasible reconciliation adjudicates conflicting federal, state, and local provisions with an auditable trace. A controlled ablation over six locally run open-weight models holds the language interface fixed and varies only the reasoning substrate, showing that reliability follows execution and not retrieval. Per-capability experiments score every operation against executable ground truth, with each accepted repair and synthesized design verified by re-execution through the core. The framework is realized over 319 rule entries from nine authorities and released as open source.

  • Top Cited Papers

    V2X-VLM: End-to-End V2X cooperative autonomous driving through large vision-Language models

    by Junwei You, Zhuoyu Jiang, Zilin Huang, Haotian Shi, Rui Gan, Keshu Wu, Xi Cheng, Xiaopeng Li, Bin Ran


  • Open Access

    Evaluating Impacts of Public Transit and Automobiles During Connected and Automated Vehicle Adoption

    by Xinmeizi Cai, Hang Zhou, Chengyuan Ma, Xiaopeng Li, Bin Ran

    The adoption of connected and automated vehicles (CAVs) and automated vehicles (AVs) is expected to improve overall travel efficiency in traffic networks. However, this improvement may vary across multimodal road traffic, influencing the equity of travel experiences among different income groups. This study evaluates the equity effects of AV deployment during its gradual adoption in public transit and automobiles. Five scenarios with varying penetration rates are designed to represent the transition from HVs to CAVs in public transit and automobiles. Different model parameters were calibrated to represent HVs, AVs, and CAVs, and four measurements of effectiveness were developed to compare travel experiences between high‐income individuals, who primarily rely on private automobiles, and low‐income individuals, who depend on public transit. A detailed case study was conducted using real‐world road network and traffic flow data from Madison, Wisconsin. Simulation results reveal that while AV adoption improves overall traffic efficiency, including nonupgraded vehicles, it also exacerbates disparities in travel performance between high‐ and low‐income groups, even with efforts to expand AV‐enabled public transit.


  • Open Access

    Distributed Connected Automated Vehicles Control Under Real‐Time Aggregated Macroscopic Car Following Behavior Estimation Based on Deep Reinforcement Learning

    by Haotian Shi, Danjue Chen, Nan Zheng, Xin Wang, Yang Zhou, Bin Ran

    This paper proposes an innovative distributed longitudinal control strategy for connected automated vehicles (CAVs) in the mixed traffic environment of CAV and human‐driven vehicles (HDVs), incorporating high‐dimensional platoon information. For mixed traffic, the traditional CAV control method focuses on microscopic trajectory information, which may not be efficient in handling the HDV stochasticity (e.g., long reaction time; various driving styles) and mixed traffic heterogeneities. Different from traditional methods, our method, for the first time, characterizes consecutive HDVs as a whole (i.e., AHDV) to reduce the HDV stochasticity and utilize its macroscopic features to control the following CAVs. The new control strategy takes advantage of platoon information to anticipate the disturbances and traffic features induced downstream under mixed traffic scenarios and greatly outperforms the traditional methods. In particular, the control algorithm is based on deep reinforcement learning (DRL) to fulfill car‐following control efficiency and further address the stochasticity for the aggregated car following behavior by embedding it in the training environment. To better utilize the macroscopic traffic features, a general platoon of mixed traffic are categorized as a CAV‐HDVs‐CAV pattern and described by corresponding DRL states. The macroscopic traffic flow properties are built upon the Newell car‐following model to capture the characteristics of aggregated HDVs' joint behaviors. Simulated experiments are conducted to validate our proposed strategy. The results demonstrate that the proposed control method has outstanding performances in terms of oscillation dampening, eco‐driving, and generalization capability.

  • Open Access

    Theoretical Research on Longitudinal Profile Design of Superhighways

    by Yongming He, Yuting Song, Yulong Pei, Bin Ran, Jia Kang

    To improve driving safety on superhighways, longitudinal profile design parameters of a superhighway are calculated via force analysis while a car is driven on a slope. The calculations consider characteristics of drivers, cars, and roads. According to the vehicle type, design speed, and natural conditions, the maximum longitudinal slope of a superhighway is calculated and compared with those of an ordinary superhighway and high-speed railway. Based on analysis of the vehicle climbing performance, braking performance, and driver visual characteristics, the maximum and minimum slope lengths of a superhighway are calculated. By analyzing the elements of vertical curves, the minimum radius and minimum length of the vertical curves of a superhighway are calculated by considering factors such as mitigating the impact at the slope bottom, driving at night, and driving time along vertical curves. Analysis and calculation results show that when the maximum longitudinal slope is 2.50%, 2.25%, and 2.00%, the minimum slope length is 450 m, 400 m, and 350 m, respectively, and the minimum vertical curve length is 145 m, 130 m, and 115 m, respectively, and the superhighway travel requirements can be satisfied at speeds of 180 km/h, 160 km/h, and 140 km/h, respectively.