A5001168378- Traffic Prediction and Management Techniques
- Traffic control and management
- Autonomous Vehicle Technology and Safety
- Simulation Techniques and Applications
- Transportation Planning and Optimization
Nanyang Technological University
2021-2024
With the increasing popularity of Digital Twin, there is an opportunity to employ deep learning models in symbiotic simulation system. Symbiotic can replicate multiple what-if instances from its real-time reference (base simulation) for short-term forecasting. Hence, it a useful tool just-in-time decision making process. Recent trends on studies emphasize combination with machine learning. Despite success and usefulness, very few works focus application such hybrid system microscopic traffic...
Symbiotic simulation systems that incorporate data-driven methods (such as machine/deep learning) are effective and efficient tools for just-in-time (JIT) operational decision making. With the growing interest on Digital Twin City, such ideal real-time microscopic traffic simulation. However, learning-based models heavily biased towards training data could produce physically inconsistent outputs. In terms of simulation, this lead to unsafe driving behaviours causing vehicle collisions in As...
In microscopic traffic simulation, it is apparent that there a dilemma between physics-based and learning-based models for modelling car-following behaviours. The former can offer analytical insights with low simulation accuracy while the latter functions like black-box, but offers high accuracy. Thus, new perspective on combining (CFM) methods given in this paper by integrating two approaches through "model calibration". CFM calibration formulated as sequential decision-making process via...
Recent works on data-driven car-following modeling have shown that a Graph Convolutional Network (GCN)-based model (CFM) can outperform other models such as Long Short-term Memory (LSTM) networks. Inspired by this result, new physics-guided GCN-based CFM is proposed in paper. The has been extended from the previous integrating vehicle platoon graph along with refinements. Furthermore, first attempt to develop graph-learning-based made paper adopting machine learning (PGML) different aspects....