ICITE2024 Invited Speakers

Prof. Zhiyuan Liu | 刘志远教授

Southeast University, China | 东南大学



Dr Zhiyuan (Terry) Liu currently a Professor in the School of Transportation at Southeast University (SEU), China; and also the Vice Dean of Graduate School of Elite Engineers in SEU. Hs is an Adjunct Professor at Monash University, Australia. He received his PhD degree from National University of Singapore (NUS) in 2011. From 2012 to 2015, he was a lecturer in Monash University, Australia. In 2018, he was a visiting scholar in the School of Mathematics and Statistics, University of Melbourne. His research interests include Transportation Data Analytics, Transportation Network Modelling, Smart Mobility, etc. In these areas, he has published more than 100 SCI/SSCI papers (all as first or corresponding author). He is an Associate Editor of several prestigious journals, including Transportation Research Part E, IEEE Intelligent Transportation Magazine, IET Intelligent Transport System and ASCE Journal of Transportation Engineering and the Managing Editor of a newly established journal with Elsevier, Multimodal Transportation.

Title of Speech: A parallel computing framework for large-scale microscopic trafficsimulation based on spectral partitioning

Abstract: Microscopic traffic simulation is a method that uses computer technology to simulate the operation of traffic systems. By creating a virtual traffic environment that includes elements such as road networks, vehicles, pedestrians, and traffic signals, it simulates real-world traffic flow and interactions. Running and analyzing traffic models can predict traffic flow, assess congestion, evaluate the impact of various traffic policies and planning schemes, and optimize traffic system management. Although microscopic simulation offers high precision and accuracy, it struggles to handle large-scale traffic networks due to computational resource limitations. With the growing demand for rapid validation and execution of large-scale network coordination and congestion mitigation strategies, the need for larger and more efficient microscopic simulations is increasing. With advancements in fields such as mathematics and computer science, parallel computing and graph theory has gradually become a research focus for improving simulation speed. We focus on how to efficiently partition the entire road network and decompose the traffic simulation into multiple sub-simulations running on different processes. This study proposes a parallel computing framework based on the Spectral Partitioning (SP) method designed to enhance the computational efficiency of large-scale microscopic traffic simulation (LSMTS). The framework employs the SP method to partition road networks, taking into account vehicle information and road information as constitutive components for node weight determination. Micro-simulation relies on vehicle information from both preceding and following vehicles to accurately infer the operational states of a vehicle. However, network partitioning can disrupt the flow of vehicle information, resulting in its loss. To address this, the proposed framework incorporates a boundary transmission method to ensure simulation accuracy and precision.

Prof. Zhigang Xu | 徐志刚教授

信息工程学院副院长

Chang’an University, China | 长安大学



Xu Zhigang, Vice Dean of the School of Information Engineering at Chang'an University, is a Level-2 professor and doctoral supervisor. He is recognized as a leading talent in the National High-level Talents Special Support Plan, a young and middle-aged technology innovation leader in the Ministry of Transport, and has been honored with the title of Outstanding Youth in Shaanxi Province, as well as awards including the Shaanxi Provincial Youth Science and Technology Award and the China Transportation Education Excellent Teacher Award. His research focuses on intelligent transportation system analysis, Internet of Vehicles and autonomous driving, and vehicle-infrastructure cooperation. He has won one second prize for National Scientific and Technological Progress, three first prizes and one third prize for Shaanxi Province Science and Technology Awards.

Title of Speech: Key Technologies of Dedicated Lanes for Connected and Automated Vehicles

Abstract: The rapid development of Connected and Automated Vehicles (CAVs) offers promising solutions to modern transportation challenges, including reducing traffic congestion, enhancing road safety, improving mobility, and lowering energy consumption. However, the integration of CAVs with traditional Human-Driven Vehicles (HDVs) on shared roads presents significant technical and operational challenges. To address these issues, the concept of Dedicated Lanes (DLs) for CAVs has emerged as an innovative strategy to improve traffic flow and enable efficient vehicle platooning.
This presentation focuses on the key technologies and strategies about DLs for CAVs. It begins by highlighting the advantages of CAVs and the role of dedicated lanes in minimizing traffic conflicts with HDVs. A comprehensive simulation platform designed to optimize DLs is introduced, combining traditional traffic simulation tools with a cyber-physical system approach. This platform supports large-scale traffic flow simulations, incorporating V2V and V2X communications to enhance visualization and realism. The second part of the presentation explores a Monte Carlo Tree Search (MCTS)-based platoon forming strategy for CAVs. This strategy allows CAVs to form closely spaced platoons while maintaining safe speeds, using cooperative lane-change decisions to achieve efficient platoon organization even in mixed traffic environments. The presentation also delves into a decentralized multi-vehicle motion planning model to optimize platoon formation. Finally, the lane-change strategies for CAV platoons in DLs are discussed. Special attention is given to managing the challenges of coordinating multiple vehicles during lane changes, especially when merging into general-purpose lanes from dedicated lanes.
This academic report outlines the latest advancements in dedicated lane technologies for CAVs, providing insights into the future of smart transportation systems and autonomous vehicle integration on public roads.