PROJECT TITLE :

A Modular Implementation Scheme for Nonsingleton Type-2 Fuzzy Logic Systems With Input Uncertainties

ABSTRACT:

Using nonsingleton (NS) input fuzzifiers along with type-two fuzzy systems to handle uncertainties in inputs has received some attention lately. But, the NS fuzzification schemes proposed thus so much have a limited impact as a result of they're restricted to a specific quite fuzzy sets only. This paper proposes a modular implementation scheme for NS sort-two fuzzy logic systems with input uncertainties. The proposed implementation scheme constitutes a generalized fuzzification that is independent of the forms/shapes of the fuzzy sets, i.e., both the NS fuzzifiers and the membership functions. To research the effectiveness of the proposed scheme, sort-a pair of fuzzy logic controllers for 3 different applications, airplane altitude control, obstacle avoidance for a mobile robot, and a wall following robot, are developed. Additionally, for mapping NS fuzzifiers for real sensors, a sensory noise pattern recognition stage is additionally developed. Despite encountering various types of membership functions and input uncertainties along these 3 applications, the proposed theme is able to successfully pander to all 3 applications. Moreover, the performance results in all three application setups show that NS fuzzification can improve the robustness of kind-a pair of fuzzy logic systems against input uncertainties.