The paper discusses a real-time traffic-adaptive signal control system refe
rred to as RHODES. The system takes as input detector data for real-time me
asurement of traffic flow, and "optimally" controls the flow through the ne
twork. The system utilizes a control architecture that (1) decomposes the t
raffic control problem into several subproblems that are interconnected in
an hierarchical fashion, (2) predicts traffic flows at appropriate resoluti
on levels (individual vehicles and platoons) to enable pro-active control,
(3) allows various optimization modules for solving the hierarchical subpro
blems, and (4) utilizes a data structure and computer/communication approac
hes that allow for fast solution of the subproblems, so that each decision
can be downloaded in the field appropriately within the given rolling time
horizon of the corresponding subproblem. The RHODES architecture, algorithm
s, and its analysis are presented. Laboratory test results, based on implem
entation of RHODES on simulation models of actual scenarios, illustrate the
effectiveness of the system. (C) 2001 Elsevier Science Ltd. All rights res
erved.