no code implementations • 3 Mar 2023 • Xuan Di, Rongye Shi, Zhaobin Mo, Yongjie Fu
For its robust predictive power (compared to pure physics-based models) and sample-efficient training (compared to pure deep learning models), physics-informed deep learning (PIDL), a paradigm hybridizing physics-based models and deep neural networks (DNN), has been booming in science and engineering fields.
no code implementations • 19 Jun 2022 • Zhaobin Mo, Yongjie Fu, Xuan Di
This paper aims to quantify uncertainty in traffic state estimation (TSE) using the generative adversarial network based physics-informed deep learning (PIDL).
1 code implementation • 19 Jun 2022 • Zhaobin Mo, Yongjie Fu, Daran Xu, Xuan Di
TrafficFlowGAN adopts a normalizing flow model as the generator to explicitly estimate the data likelihood.
no code implementations • 6 Jun 2021 • Rongye Shi, Zhaobin Mo, Kuang Huang, Xuan Di, Qiang Du
Traffic state estimation (TSE) bifurcates into two categories, model-driven and data-driven (e. g., machine learning, ML), while each suffers from either deficient physics or small data.
no code implementations • 17 Jan 2021 • Rongye Shi, Zhaobin Mo, Kuang Huang, Xuan Di, Qiang Du
This paper focuses on highway TSE with observed data from loop detectors, using traffic density as the traffic variables.
no code implementations • 24 Dec 2020 • Zhaobin Mo, Xuan Di, Rongye Shi
We design physics-informed deep learning car-following (PIDL-CF) architectures encoded with two popular physics-based models - IDM and OVM, on which acceleration is predicted for four traffic regimes: acceleration, deceleration, cruising, and emergency braking.
no code implementations • 28 Feb 2018 • Sisi Li, Wenshuo Wang, Zhaobin Mo, Ding Zhao
Learning knowledge from driving encounters could help self-driving cars make appropriate decisions when driving in complex settings with nearby vehicles engaged.
no code implementations • 27 Feb 2018 • Zhaobin Mo, Sisi Li, Diange Yang, Ding Zhao
To overcome this problem, we extract naturalistic V2V encounters data from the database, and then separate the primary vehicle encounter category by clustering.