A new approach for fuzzy modeling using input-output data with data uncertainty

This paper proposes a new approach for fuzzy modeling using numerical data with data uncertainty, which, is named as a robust fuzzy tree approach. The data for modeling are subjected to data uncertainty with a known structure and unknown bat bounded data uncertainty respectively. The suggested algorithms employ an adaptive partition method-based on a binary tree to determine the number of rules and the respective premise parts. And the optimal consequent parameters are obtained by solving a set of robust linear estimation problems. Finally, simulations show the validity of the method.