About

Our lab investigates the nervous systems of octopuses and bees, two ideal systems for studying end-of-life transitions and death from evolutionary and developmental perspectives. We use multiple high-dimensional omics, behavioral, and molecular approaches to uncover fundamental rules of aging, senescence, and death.

Research topics

• Artificial Intelligence
• Computer Science
• Data Mining
• Engineering
• Machine Learning
• Transport engineering
• Real-time computing
• Embedded system
• Automotive engineering
• Simulation
• Aerospace engineering
• Control engineering
• Structural engineering
• Computer network

Selected publications

• Prof. Yinhai Wang [ITS People]

  IEEE Intelligent Transportation Systems Magazine · 2026-05-01

  article1st authorCorresponding
  PublisherDOI

• EditFollower: Tunable Car Following Models for Customizable Driving Behavior

  IEEE Transactions on Intelligent Transportation Systems · 2025-05-16 · 1 citations

  articleSenior author

  In the realm of driving technologies, fully autonomous vehicles have not been widely adopted yet, making advanced driver assistance systems (ADAS) crucial for enhancing driving experiences. Among these, car-following behavior modeling plays a pivotal role, forming the foundation for systems that ensure safe and efficient vehicle interactions. However, current approaches often rely on fixed parameters, failing to capture the diverse social preferences and driving styles of individuals. To overcome these limitations, we propose the Editable Behavior Generation (EBG) model, a data-driven car-following model that allows for adjusting driving discourtesy levels. The framework integrates diverse courtesy calculation methods into long short-term memory (LSTM) and Transformer architectures, offering a comprehensive approach to capture nuanced driving dynamics. By integrating various discourtesy values during the training process, our model generates realistic agent trajectories with different levels of courtesy in car-following behavior. Experimental results on the naturalistic datasets showcase a reduction in Mean Squared Error (MSE) of spacing and MSE of speed compared to baselines, establishing style controllability. To the best of our knowledge, this work represents the first data-driven car-following model capable of dynamically adjusting discourtesy levels. Our model provides valuable insights for the development of ADAS that take into account drivers’ social preferences.

  PublisherDOI

• Engaging High School Students in a DOT-Funded Summer Camp to Promote Transportation Engineering Majors and Careers

  2025-08-21

  articleSenior author
  PublisherDOI

• Deep Fictitious Play-Based Potential Differential Games for Learning Human-Like Interaction at Unsignalized Intersections

  ArXiv.org · 2025-06-14

  preprintOpen accessSenior author

  Modeling vehicle interactions at unsignalized intersections is a challenging task due to the complexity of the underlying game-theoretic processes. Although prior studies have attempted to capture interactive driving behaviors, most approaches relied solely on game-theoretic formulations and did not leverage naturalistic driving datasets. In this study, we learn human-like interactive driving policies at unsignalized intersections using Deep Fictitious Play. Specifically, we first model vehicle interactions as a Differential Game, which is then reformulated as a Potential Differential Game. The weights in the cost function are learned from the dataset and capture diverse driving styles. We also demonstrate that our framework provides a theoretical guarantee of convergence to a Nash equilibrium. To the best of our knowledge, this is the first study to train interactive driving policies using Deep Fictitious Play. We validate the effectiveness of our Deep Fictitious Play-Based Potential Differential Game (DFP-PDG) framework using the INTERACTION dataset. The results demonstrate that the proposed framework achieves satisfactory performance in learning human-like driving policies. The learned individual weights effectively capture variations in driver aggressiveness and preferences. Furthermore, the ablation study highlights the importance of each component within our model.

  PublisherOA PDFDOI

• Road Side Unit Location Optimization Considering Communication Channel Competition and 6G Technology

  IEEE Transactions on Intelligent Transportation Systems · 2025-06-02 · 3 citations

  article

  This study investigates the problem of road side unit (RSU) location optimization considering vehicle-to-RSU (V2R) communication channel competition. To hedge against the uncertainty of vehicle density, the problem is formulated as a stochastic mixed-integer nonlinear program with equilibrium constraints. This program aims to minimize the expectation of weighted sum of V2R communication delay, packet loss rate and packet collision rate and age of information in V2R communication over all scenarios given RSU location budget limit. Decision variables are RSU locations and the number of connected autonomous vehicles (CAVs) communicating with each located RSU. Equilibrium constraints in the program model V2R communication channel competition among CAVs and ensures the choice of CAVs on RSUs to satisfy user equilibrium principle. The V2R communication is calculated under 6G technology. The program is linearized by using piecewise linearization method. To enhance the solution efficiency, a progressive hedging algorithm is developed to decompose the relaxed linearized model into several subproblems. The optimal solution to the relaxed linearized model is found by iteratively formulating and the solving subproblems. A branch and bound algorithm is introduced to obtain the optimal integer solution to the linearized model. The numerical results show that the proposed model can achieve 20.55% lower total communication delay than the state-of-the-art model only optimizing total V2R information propagation delay, when CAVs choose RSUs for communication in a competitive manner.

  PublisherDOI

• Large language models and their applications in roadway safety and mobility enhancement: A comprehensive review

  Artificial Intelligence for Transportation · 2025-07-01 · 13 citations

  reviewOpen accessSenior authorCorresponding
  PublisherOA PDFDOI

• Toll Programs and Tolling Equity in the USA: Current Best Practices

  Journal of Transportation Engineering Part A Systems · 2025-07-09

  articleSenior author

  Tolling is increasingly being used in the United States for various purposes, e.g., reducing congestion and other externalities of traffic, increasing roadway efficiency, and as a funding source for constructing new infrastructure, maintaining existing infrastructure, and other projects. Tolling is considered a true user fee, where roadway users more directly pay for the use of roadways compared to the classic methods of roadway fee collection, namely, gas taxes. The collection of tolls, however, raises several equity concerns, as these cost burdens have the potential to disproportionately affect low-income and vulnerable communities. Ultimately, the effectiveness of a tolling facility in mitigating these equity concerns is determined by the local context of the existing transportation network, the makeup of the surrounding communities, the existing and anticipated users of the facility, the input the local community can give regarding the toll facility, and the redistribution schemes for the collected toll fees. These local variables will determine how progressive or regressive a tolling facility is for the communities it traverses. Following these concepts, it is possible to use different equity measures to define these ideas of progressiveness and regressiveness. These equity measures can quantify the effectiveness of the process by which toll programs are implemented and maintained (process equity) and the outcomes for the transportation network users, i.e., those who do and do not use toll facilities (outcome equity). Nine existing or proposed low-income toll programs were explored to showcase the lessons learned for the most common methods for mitigating the equity concerns raised by the use of tolling facilities.

  PublisherDOI

• UAV-Based Vehicle Re-Identification via Counterfactual Attention Learning and Hard-Sensitive Binomial Joint Promotion

  IEEE Transactions on Intelligent Transportation Systems · 2025-10-20

  articleSenior author

  Autonomous Aerial Vehicles (AAVs) based Vehicle Re-Identification (ReID) brings high flexibility to the ReID system, and also brings challenges of complicated shooting views and special occlusions. In this study, we propose a novel framework for UAV-based vehicle ReID, called <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Counterfactual Attention and Joint Promotion based ReID</i> (CAJP-ReID), to deal with fine-grained feature extraction and occlusions. Based on the mechanism of using randomly generated counterfactual attention intervention to train, the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Counterfactual Attention Learning Network</i> (CAL-Net) is proposed to learn the fine-grained features of vehicle images, for distinguishing similar vehicles in different ID categories. In order to enhance the diversity of the dataset and the robustness of the network to occluded images, the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Counterfactual Attention Enhancement Module</i> (CAE-Module) is proposed based on counterfactual attention mechanism, by the data-enriching mechanism of cropping and erasing. The <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Hard-sensitive Binomial Joint Promotion Loss</i> (HBJP-Loss) is proposed to comprehensively consider the relative distance and absolute distance of positive and negative samples of vehicle images, and further improves the accuracy of vehicle re-identification. Experiments on two public datasets show that the proposed method achieves State-of-the-Art performance.

  PublisherDOI

• Communication-aware Diffusion Models for Multi-Agent Trajectory Forecasting in Connected and Autonomous Vehicles

  2025-12-03

  articleOpen accessSenior author

  Trajectory prediction plays a critical role in connected and autonomous vehicles (CAVs), particularly in scenarios involving multiple interacting agents. In such dynamic settings, where CAVs and roadside units (RSUs) communicate to share observations and predictions, accurately forecasting the trajectories of all involved agents becomes even more critical. In this work, we introduce a communication-aware diffusion model that leverages collaborative communication among CAVs during inference. Unlike previous approaches that modify the training objective, our method applies the guidance only during inference. By adding three different communication-aware guidances, we reduce displacement errors and improve accuracy without modifying the training pipeline. We validate our method on the INTERACTION dataset and demonstrate better performance with the introduction of guidance in terms of ADE and FDE. Furthermore, our approach is modular and computationally efficient, making it particularly well-suited for real-time deployment in edge-assisted CAV systems where robust and low-latency predictions are critical. This work highlights how communication-aware guidance can bridge the gap between cutting-edge generative models and the system-level demands of multi-agent CAV applications.

  PublisherOA PDFDOI

• Is cooperative always better? Multi-Agent Reinforcement Learning with explicit neighborhood backtracking for network-wide traffic signal control

  Transportation Research Part C Emerging Technologies · 2025-07-25 · 31 citations

  articleSenior author
  PublisherDOI

Frequent coauthors

• Timothy V. Larson

  University of Washington

  101 shared

• Mohammed Saad

  University Hospital Schleswig-Holstein

  100 shared

• L J. Sally Liu

  Institute of Social and Preventive Medicine

  100 shared

• Daniel OʼLeary

  Tufts University

  100 shared

• Robert Detrano

  University of California, Irvine

  100 shared

• Lianne Sheppard

  University of Washington

  100 shared

• Thomas Lumley

  100 shared

• Timothy Nyerges

  University of Washington

  100 shared

Labs

• ZYWang LabPI

Awards & honors

• Allen Institute Next Generation Leader and National Postdoct…