FUZZY EXPERT SYSTEMS VS. NEURAL REVISITED NETWORKS - TRUCK BACKER-UPPER CONTROL
Prabhu Ramamoorthy, Song Huang · 1991
&search on neural networks and fuzzy logic have progressed on two independent paths. In general, fuzzy logic uses verbal information for handling higher-order logical relations between inputs and outputs which are not crisply defined. On the other hand, neural networks are used to obtain information about systems from large input/output observations and training or learning procedures. From these definitions, it appears that fuzzy logic and neural network fulfill two complementary functions. Hence, a merger of these two concepts could lead to powerful yet flexible knowledge processing tools. This paper provides some insights along these lines using the truck-backer-upper control problem. New network architectures by merging these two concepts and simulation results for the truck-back-upper problem using the new architecture are also shown in this paper.