Coordinated Control of Multiple Ramp Metering Based on DHP (¿) Controller

Abstract

we presented a novel control method in previous studies, which is action-dependent heuristic dynamic programming based on eligibility traces (ADHDP (lambda)). Although ADHDP (lambda) has good performance in local ramp metering control, it can't efficiently deal with the problems of multiple ramp metering control. Then in this article, a new method which is called dual heuristic programming based on eligibility traces (DHP (lambda)) is presented. The new method DHP (lambda) is composed of DHP and eligibility traces. DHP has efficiently solved the problem of coordinate ramp metering control, and eligibility traces can accelerate the learning rate of DHP. DHP (lambda) is an approximate optimal ramp metering method. Simulation studies on a hypothetical freeway are reported, whereby the results are compared with those obtained by applying ALINEA as a stand-alone strategy. It is shown that the proposed control scheme is efficient.