CMAC modeling using bacterial evolutionary algorithm (BEA) on field programmable gate array (FPGA)

Evolutionary algorithms (EAs) have been applied to various combinatorial optimization problems. Bacterial evolutionary algorithm (BEA) is an optimization method that incorporates special mechanisms inspired by natural phenomena of microbial evolution. The BEA has been applied to fuzzy modeling of nonlinear systems for efficient discovery of appropriate rules and parameter tunings. CMAC is a neural network that imitates human cerebellum. One of the advantages of CMAC is its fast learning capability. It is, however, difficult to identify appropriate parameters of CMAC. This paper proposes a new approach to determination of input space division of CMAC using BEA. CMAC learning using BEA needs a large amount of computation time. This paper presents a pipelined BEA/CMAC hardware that is about 96 times faster than that by the software executed on the conventional processor. This paper also presents a new approach to CMAC modeling by encoding granule cells of CMAC into chromosomes of BEA for identification of more accurate model