Computational intelligence in control-a comparison of several neuro-fuzzy systems

Abstract

The aim of this paper is to present and compare several neuro-fuzzy systems used as controllers for simulated backing up of a truck to a loading dock in a planar parking lot. The following systems have been considered: nfCon-the system proposed in this paper, an alternative neuro-fuzzy system reported in literature as well as the well-known ANFIS and NFIDENT systems. The main criterion of comparison of all systems is their performance (the accuracy of functioning) versus interpretability the transparency and the ability to explain generated actions; it also includes an analysis and pruning of the obtained fuzzy-rule bases.