An intelligent control approach to decentralized routing and flow control in highways

Abstract

A decentralized intelligent controller for routing in multi-destination transportation networks is presented. The controller is based on a dynamic congestion measure which was recently derived from the solution of an optimal control problem. The proposed controller is decentralized in the sense that all the on-line computations are done locally at the nodes with minimum information transfer from the adjacent downstream nodes. Furthermore, no synchronization among the nodes is needed. The controller avoids congestion on the links by not allowing any traffic higher than the capacity of a given link. Thus the controller also acts as a flow controller. An example is also presented to illustrate the design methodology and to assess the controller's performance in a typical case.