Road congestion is a complex, significant and continuing problem and is a major factor in reduced air quality. This project is concerned with urban road systems and traffic signal control. The work presents the algorithms employed by experimental traffic signals used in a simulated road network. The network is a grid system that is specifically contrived to exhibit highly interactive tidal congestion. The available forward road capacity (AFRC) algorithm is presented as an extension of existing MOVA control algorithms. The objective of providing maximum junction utilisation under congested conditions is explained along with the proposed solution. AFRC automatically determines the temporal and geographic boundaries of congestion and attempts to resolve the problem across multiple junctions without the need for prior knowledge of traffic flows. The system is therefore self-organising and, as a result, is able to respond to dynamic and unanticipated flow changes without manual reconfiguration. Simulation results are presented which compare the action of MOVA and AFRC detection in managing high density congestion across the multiple junction network. The results obtained show that delays can be significantly reduced in periods of congestion and that the performance improvement increases as the road becomes more congested. Reducing overall congestion duration generates time savings for individuals and a reduction in the negative effects on the environment and public health, increased fuel efficiency and reduced fuel waste. The main benefits of the system are equally applicable to all powered transport systems regardless of fuel source or control method including autonomous vehicles and vehicles equipped with V2X communication. The controller is an ongoing development and is capable of directly and accurately measuring the degree of congestion providing metrics that may be used for other controls including gating and active traffic guidance.
Available Forward Road Capacity Detection Algorithms to Reduce Urban Traffic Congestion
01.06.2017
574030 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Non-recurrent traffic congestion detection on heterogeneous urban road networks
Taylor & Francis Verlag | 2015
|Earliest-Deadline-Based Scheduling to Reduce Urban Traffic Congestion
Online Contents | 2014
|