Coordinated traffic-responsive ramp control via nonlinear state feedback

In this paper, we develop a coordinated traffic responsive ramp control str ategy based on feedback control and artificial neural networks. The propose d feedback control law is nonlinear and realized by a series of neural netw orks. The parameters of the neural networks are obtained through a nonlinea r optimization procedure. Traffic simulations show that the proposed nonlin ear ramp control strategy compares favorably against the well-known linear quadratic (LQ) control strategy in reducing total travel times, particularl y at situations where drastic changes in traffic demand and road capacity o ccur. (C) 2001 Elsevier Science Ltd. All rights reserved.