Parallel Inference Method of Fuzzy Rules

Abstract

During the past decade, several specific hardware systems for fast fuzzy inference have been developed. However, most of them are dedicated to be executed sequentially rule by rule, although they pose parallel processing manners in operations of Min/Max computations. In this paper, we propose a novel parallel inference method which is capable of processing several fuzzy rules simultaneously by using a T800 Transputer system. For this, each membership function has its own index. Only active rules are searched and executed concurrently. We also analyze the worst case upper bound for internode message delays. The proposed system has a relatively higher speedup, compared with the results of experiments in a single processor. This parallel fuzzy inference system can be used for a system requiring a rapid inference time in a real-time system. This system is applied to build powerful architectures for control applications such as robot control.