A brief comparison of fuzzy associative memory models for guiding autonomous problems

Abstract

Fuzzy associative memories (FAMs) are models inspired in the human brain ability to store and recall information. These models can be used for the storage of associations of fuzzy sets and, thus, they can be used as inference engines in fuzzy controllers. Several FAM models have been developed in the last years, but we are not aware of a work comparing the performance of novel FAMs in control. In this paper, we briefly investigate the performance of some FAMs in the automatic guidance problems of backing-up a truck (BT) and backing-up a truck and trailer (BTT). In particular, we note that the dual implicative fuzzy associative memories (co-IFAMs) constitute an interesting alternative to traditional models such as that of Kosko and Mamdani.

Keywords

• Fuzzy set theory
• fuzzy associative memory
• inference engine
• fuzzy controller