Traffic network hybrid feedback controller via 0-1 classification of PWARX system with hierarchy

A new design method for the traffic network hybrid feedback controller is proposed. In the proposed method, the piece-wise auto regressive exogenous (PWARX) classifier describes a nonlinear feedback control law of the traffic control system, where the output of a previously developed controller is reproduced applying the 0-1 classifications of the PWARX system. A new performance criterion is presented here to consider directly the covariance of the traffic network controller output y. The proposed method is a hierarchical classification procedure, where the cluster splitting process is introduced to the cluster with the worst classification performance (which includes 0-1 mixed values of y). The cluster splitting process is followed by the piecewise fitting process to compute the cluster guard and dynamics, and the cluster updating process to find new centre points of the clusters. The usefulness of the proposed method is verified through some numerical experiments.