Agent Based Real-time Signal Coordination in Congested Networks
www.purdue.edu/discoverypark/nextrans/research/research_in_progress.php
Record Type: UTC
This study is a continuation of an ongoing NEXTRANS study on agent-based reinforcement learning methods for signal coordination in congested networks. The ongoing study is a joint effort between University of Illinois and Purdue University, where each intersection is controlled by an agent and some information is shared between adjacent agents. Researchers will aim to develop a framework to study signal coordination and real-time adaptive routing within an agent based model. The agent based approach has clear advantages over more traditional approaches as the learning method does not require modeling the system and the operation is free of restrictions imposed by cycle-based strategies.
Start date: 2012/8/1
Status: Active
Contract/Grant Number: DTRT12-G-UTC05
Total Dollars: 209602
Source Organization: Purdue University, West Lafayette
Notes: University of Illinois at Urbana-Champaign; Purdue University
Date Added: 07/12/2012
Index Terms: Traffic signal control systems, Real time control, Agent based models, Adaptive control, Routing,
|
|
NEXTRANS
http://www.purdue.edu/discoverypark/nextrans/
Purdue University, West Lafayette
3000 Kent Avenue
Lafayette, IN 47906
USA
Phone: (765) 496-9729
Fax: (765) 807-3123
|
|
Benekohal, Rahim F.
Phone: (217) 244-6288
Email: rbenekoh@uiuc.edu
Ukkusuri, Satish
|
