Search

The purpose of signal control is to allocate time for competing traffic flows to ensure safety. Artificial intelligence has made transportation researchers more interested in adaptive traffic signal control, and recent literature confirms that deep reinforcement learning (DRL) can be effectively applied to adaptive traffic signal control. Deep neural networks enhance the learning potential of reinforcement learning. This study applies the DRL method, Double Deep Q‐Network, to train local agents. Each local agent learns independently to accommodate the regional traffic flows and dynamics. After completing the learning, a global agent is created to integrate and unify the action policies selected by each local agent to achieve the purpose of traffic signal coordination. Traffic flow conditions are simulated through the simulation of urban mobility. The benefits of the proposed approach include improving the efficiency of intersections and minimizing the overall average waiting time of vehicles. The proposed multi‐agent reinforcement learning model significantly improves the average vehicle waiting time and queue length compared with the results from PASSER‐V and pre‐timed signal setting strategies. [ABSTRACT FROM AUTHOR]

The focus of this research project was to develop a new simulation framework for analyzing traffic networks with Advanced Traveler Information Systems (ATIS) and/or Advanced Traffic Management Systems ( ATMS). The report describes the development of DYNASMART (DYnamic Network Assignment-Simulation Model for Advanced Road Telematics). The report is presented as follows: Chapter 1 discusses the simulation approach of DYNASMART and explains the traffic control features incorporated in it during the research. Chapter 2 discuss the simulations performed with a trial network with ATMS controls. Chapter 3 concentrates on the ATIS simulations on the Anaheim network. Results from the simulations of traffic management for special-events traffic from the Anaheim stadium are presented in Chapter 4, followed by overall conclusions.

This research aims at developing a system of real-time simulation-based dynamic traffic model under mixed traffic flow conditions. The system consists of two layers, namely, simulation layer and real-time control layer. The system is implemented based on rolling horizon approach, which is advanced for each stage; thus real-time data can be incorporated within the framework. In order to predict normative, as well as predictive, information, a simulation-based dynamic traffic assignment model is employed within each stage. Empirical data for a real city network, such as flows from vehicle detectors, are used to validate the model in a real- time environment. The values of mean absolute percentage error and root-mean-squared percentage error are within 15%, and the results show promising agreements between observed and simulated flows. DOI: 10.1061/(ASCE)TE.1943-5436.0000168 CE Database subject headings: Time dependence; Estimation; Traffic assignment; Traffic models. Author keywords: DynaTAIWAN; Real-time deployment; Time-dependent O-D estimation.

Developing demand responsive transit systems are important with regard to meeting the travel needs for elderly people. Although Dial-a-ride Problems (DARP) have been discussed for several decades, most researchers have worked to develop algorithms with low computational cost under the minimal total travel costs, and fewer studies have considered how changes in travel time might affect the vehicle routes and service sequences. Ignoring such variations in travel time when design vehicle routes and schedules might lead to the production of inefficient vehicle routes, as well as incorrect actual vehicle arrival times at the related nodes. The purpose of this paper is to construct a DARP formulation with consideration of time-dependent travel times and utilizes the traffic simulation software, DynaTAIWAN, to simulate the real traffic conditions in order to obtain the time-dependent travel time matrices. The branch-and-price approach is introduced for the time-dependent DARP and tested by examining the sub-network of Kaohsiung City, Taiwan. The numerical results reveal that the length of the time window can significantly affect the vehicle routes and quantitative measurements. As the length of the time window increases, the objective value and the number of vehicles will reduce significantly. However, the CPU time, the average pickup delay time, the average delivery delay time and the average actual ride time (ART)/direct ride time (DRT) will increase significantly as the length of the time window increases. Designing the vehicle routes to reduce operating costs and satisfy the requirements of customers is a difficult task, and a trade-off must be made between these goals. Copyright © 2014 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]