A5058410192- Autonomous Vehicle Technology and Safety
- Traffic control and management
- Traffic Prediction and Management Techniques
- Transportation Planning and Optimization
- Traffic and Road Safety
- Vehicle emissions and performance
- Robotic Path Planning Algorithms
- Transportation and Mobility Innovations
- Human-Automation Interaction and Safety
- Advanced Neural Network Applications
- Air Quality and Health Impacts
- Human Mobility and Location-Based Analysis
- Air Quality Monitoring and Forecasting
- Data Management and Algorithms
- Video Surveillance and Tracking Methods
- Robotics and Sensor-Based Localization
- Topic Modeling
- Simulation Techniques and Applications
- Vehicle Dynamics and Control Systems
- Industrial Vision Systems and Defect Detection
- Vehicular Ad Hoc Networks (VANETs)
- Infrastructure Maintenance and Monitoring
- Urban Transport and Accessibility
- Safety Warnings and Signage
- Reinforcement Learning in Robotics
Hong Kong University of Science and Technology
2023-2025
University of Hong Kong
2023-2025
Southeast University
2025
University of Washington
2018-2023
Iowa Department of Transportation
2023
United States Department of Transportation
2023
ORCID
2021
Ministry of Education of the People's Republic of China
2016-2020
National Transportation Research Center
2020
Oak Ridge National Laboratory
2019-2020
Abstract Ride comfort plays an important role in determining the public acceptance of autonomous vehicles (AVs). Many factors, such as road profile, driving speed, and suspension system, influence ride AVs. This study proposes a hierarchical framework for improving by integrating speed planning control vehicle‐to‐everything environment. Based on safe, comfortable, efficient via dynamic programming, deep reinforcement learning‐based is proposed to adapt changing pavement conditions....
An effective forward collision warning (FCW) system must be compatible with drivers' risk perceptions and behavioral responses. The Collision Avoidance Metrics Partnership (CAMP) developed a kinematic-based FCW algorithm to determine the minimum distance needed stop safely under various levels of rear-end crash risk. generates linear function for predicting expected response decelerations (ERDs) by considering motions involved vehicles. This works well when risks perceived drivers are low;...
Car-following is the most common driving scenario where a following vehicle follows lead in same lane. One crucial factor of car-following behavior style which affects speed and gap selection, acceleration pattern, fuel consumption. However, existing research used limited categories through pre-defined patterns failed to encode into data-driven models. To address these limitations, we propose Aggressiveness Informed Car-Following (AICF) modeling approach, embeds as dynamic input feature In...
Real-time public transit ridership flow and origin-destination (O-D) information is essential for improving service quality optimizing networks in smart cities. The effectiveness accuracy of the traditional survey-based methods card data-driven O-D inference have multiple disadvantages terms biased results, high latency, insufficient sample size, cost time energy. By considering ubiquity mobile devices world, monitoring can be accomplished by passively sensing Wi-Fi Bluetooth (BT)...
Traffic signal control is important for intersection safety and efficiency. However, most traffic methods are designed individual intersections or corridors. Although some adaptive systems have been developed, the used often proprietary not published, making it difficult to evaluate their effectiveness. This study proposes an multi-input multi-output method that only can improve network-wide operations in terms of reduced delay energy consumption, but also more computationally feasible than...
Customized path-based speed prediction is an eventful tool for congestion avoidance, route optimization and travel time navigation apps, cab-hailing companies autonomous vehicles. Traditionally, the algorithms are based on road segments can only support several main roads. Path-based very challenging since always changing in different path locations jointly affected by lots of complicated factors. This article presents a novel deep learning framework customized prediction. A Speed Prediction...
To build more accurate and realistic freeway car-following models, driving characteristics specific to car following should be considered. This study, therefore, analyzed three models calibrated for different styles. A total of 5,900 events were extracted from 161,055 km data collected in the Shanghai Naturalistic Driving Study (SH-NDS) database. Based on fuzzy inference system built this these categorized as representing one two styles: nonaggressive or aggressive. The styles visualized by...
Surveillance cameras are widely deployed traffic sensors, due to their affordable prices and being able capture rich information. However, current surveillance systems have not been fully exploited: these isolated can only extract information from own fixed views. To enable a collaborative sensing system, we propose novel framework called Traffic-Informed Multi-camera Sensing (TIMS) system for network-level estimation. By pushing multi-camera Re-IDentification (ReID) workflow towards...
Car-following refers to a control process in which the following vehicle (FV) tries keep safe distance between itself and lead (LV) by adjusting its acceleration response actions of ahead. The corresponding car-following models, describe how one follows another traffic flow, form cornerstone for microscopic simulation intelligent development. One major motivation models is replicate human drivers' longitudinal driving trajectories. To model long-term dependency future on historical...
With the booming of machine learning (ML) empowered autonomous vehicles (AVs), their interpretability per se is arousing more and attention. This issue plays a crucial role in public trust acceptance toward AVs. To enhance ML-empowered AVs, mainstream pipelines remain module-based architectures introduce large language models to interpret behavior In this paper, we provide brand-new avenue. Recent studies neuroscience suggest that it cognitive maps human brains grant humans highly...
The push towards sustainable transportation emphasizes vehicular energy efficiency in mixed traffic scenarios. A research hotspot is the cooperative control of connected and automated vehicles (CAVs), particularly contexts involving uncertainties human-driven (HDVs). Cooperative strategies are pivotal improving driving safety, efficiency, reducing consumption. Our study introduces a strategy for CAVs based on multi-agent twin delayed deep deterministic policy gradient (MATD3) algorithm. We...
Integrating mobile monitoring data with street view images (SVIs) holds promise for predicting local air pollution. However, algorithms, sampling strategies, and image quality introduce extra errors due to a lack of reliable references that quantify their effects. To bridge this gap, we employed 314 taxis monitor NO, NO2, PM2.5, PM10, extracted features from ∼382,000 SVIs at multiple angles (0°, 90°, 180°, 270°) buffer radii (100-500 m). Additionally, three typical machine learning...