TRB Pubsindex
Text Size:
x

« Prev Next »

Return to Results | New Search

Bookmark and Share

Title:

Enhancing Feedback Perimeter Controllers for Urban Networks by Use of Online Learning and Data-Driven Adaptive Optimization

Accession Number:

01593869

Record Type:

Component

Abstract:

An adaptive scheme for perimeter control of heterogeneous transportation networks is presented. The proposed methodology utilizes the concept of the Macroscopic Fundamental Diagram (MFD) integrated with an adaptive optimization technique. First, a new MFD-based macroscopic model is introduced to describe the dynamics of heterogeneous urban networks that can be partitioned in a small number of homogeneous regions. The non-linear model describes the evolution of the multi-region system over time assuming the existence of well-defined MFDs. Many linear approximations of the model (for different set-points) are used for designing optimal multivariable integral feedback regulators. Since the resulting regulators are derived from approximations of the non-linear model, they are further enhanced in real-time based on performance measurements and online learning/adaptive optimization. The recently proposed Adaptive Fine-Tuning (AFT), an iterative data-driven algorithm is used for that purpose and its objective is to optimize the gain matrices and set-points of the multivariable perimeter controller based on real-time observations. The derived control scheme is tested in micro-simulation and different evaluation criteria are studied. The urban network of Barcelona, Spain is partitioned in four homogeneous regions and perimeter flow control is applied in the common boundaries between regions. The simulation results show that the total delay in the network decreases significantly by only controlling a small number of intersections. It is worth noting, that since the boundaries of the network are not controlled (only internal intersections are considered) the controller achieves a better distribution of congestion between the regions, thus preventing the network degradation and improving total outflow.

Supplemental Notes:

This paper was sponsored by TRB committee AHB45 Standing Committee on Traffic Flow Theory and Characteristics.

Monograph Accession #:

01584066

Report/Paper Numbers:

16-3873

Language:

English

Corporate Authors:

Transportation Research Board

500 Fifth Street, NW
Washington, DC 20001 United States

Authors:

Kouvelas, Anastasios
Saeedmanesh, Mohammadreza
Geroliminis, Nikolas

Pagination:

20p

Publication Date:

2016

Conference:

Transportation Research Board 95th Annual Meeting

Location: Washington DC, United States
Date: 2016-1-10 to 2016-1-14
Sponsors: Transportation Research Board

Media Type:

Digital/other

Features:

Figures; References; Tables

Geographic Terms:

Barcelona (Spain)

Subject Areas:

Highways; Operations and Traffic Management

Source Data:

Transportation Research Board Annual Meeting 2016 Paper #16-3873

Files:

TRIS, TRB, ATRI

Created Date:

Jan 12 2016 5:41PM