Self-adaptation of fuzzy controller optimized by hardware-based GA

Self-adaptation of fuzzy controller optimized by hardware-based genetic algorithm is proposed in this paper. GAs are search algorithms based on the mechanics of natural selection and natural genetics. Advantageous to hardware implementation, an architecture of CA described by VLSI language is proposed in this paper which can perform the functions of population storage, selective crossover, mutation, fitness storage and survival determination. The hardware CA processor has been implemented in Altera FPGA FLEX10K40 and is a general-purpose VLSI architecture. A nonlinear control system of inverted pendulum is simulated on-line with the self-adaptive fuzzy controller whose optimization is focused on the dynamic learning of rules store by GA, which proves the validity and the applicability of the presented design.