An IPSO-based model free control for freeway ramp metering

A model free control strategy based on improved particle swarm optimization (IPSO) is proposed for freeway ramp metering. First, the freeway traffic model is built which is nonlinear in nature and can be linearized to a universal model with error correction. The control objective is to maintain the freeway operated at a desired traffic density. Then, an improved PSO algorithm is presented which has high speed convergence and can overcome the problem of being trapped in local optimum. The unknown parameters in the controller are optimized by the IPSO and the range of the particles can be given by analyzing the convergence of the controller. Finally, a simulation example is given to demonstrate the effectiveness of the proposed method.