Nagoya University · Engineering
Professor Hong Zhu's research lab specializes in intelligent transportation systems, with a focus on adaptive traffic signal control, digital twin technology, and microscopic traffic simulation. The lab investigates dynamic traffic management strategies using multi-agent reinforcement learning, cellular automata modeling, and advanced simulation frameworks to address real-world challenges such as downstream queue effects, capacity drop, and start-up lost time. Key research directions include the development of data-driven adjustment models for signal performance and the integration of digital twins for real-time, synchronized traffic control under uncertain conditions.
Figures are computed from collected data and may differ slightly.
Unmanned traffic signal control is regarded as a sustainable intelligent management methodology. However, it faces the challenge of unpredictable traffic flow due to stochastic arrivals. The digital twin (DT) has emerged as a promising approach to address the challenges of time-varying traffic demand in urban transportation. Previous studies of DT-based adaptive traffic signal control (ATSC) methods all assume ideal synchronization conditions between the DT and the physical twin (PT). It means t
The efficiency of road networks affects the daily activities of each stakeholder. Multi-agent reinforcement learning (MARL) has emerged as a method for managing network traffic signal control (TSC). It treats each intersection as an agent and coordinates their actions to enhance overall performance. A critical issue is enabling agents to appropriately and systematically respond to network demand changes. In response, this study proposes a coordination graph-based framework. It considers two adja
Arterials are important transportation facilities, undertaking the two functions of mobility and accessibility. In the urban area, signalized intersections along arterials are usually closely spaced and bear heavy traffic pressure. Capacities of intersections can be reduced by downstream intersections even without having spillback. This effect will be accumulated and amplified back along the traffic direction and may lead to severe congestion in the upstream intersection which can be frequently
Vehicular flow in highway is inherently complex and development of microscopic models of vehicular flow has been a daunting task for researchers. This paper presents the use of Cellular automata (CA) micro simulation for modeling multi-lane traffic characteristics in highway, as well as considering the average speed difference (ASD) and lane-changing rules (LCR) in the CA model. Firstly, on the base of the Symmetric Two-Lane Cellular Automata (STCA) model, we analyzed the impact on traffic chara
Start-up lost time (SLT, hereinafter) is one of the basic parameters for evaluating the capacities of signalized intersections. In the urban area, intersections are closely spaced and traffic demand is high. Drivers’ ambitions for the quick startup are demotivated by negative downstream traffic conditions, for example, limited available storage length, long queue and/or poorly coordinated traffic signals. The performance of upstream intersection is deteriorated including the growth in the SLT. I
Former studies revealed that before spillbacks happen intersection capacities can still be greatly affected by downstream queues and signals. However, there is no methodology reasonably accounting for this effect. This research aims to propose adjustment factor models (Adj models) for both saturation flow rate (SFR) and start-up lost time (SLT) in through lane groups considering downstream conditions. An improved cell transmission model is created to build a simulation platform that can well rep
Open papers in the app to read, cite, and organize with AI.
