Memristor-Based GFMM Neural Network Circuit of Biology With Multiobjective Decision and its Application in Industrial Autonomous Firefighting

Abstract

Current memristive circuits for biological decision-making only consider simple situations and do not take into account how the organisms themselves learn these behaviors. In this article, a memristor-based generalized fuzzy min–max (GFMM) neural network circuit of biology with multiobjective decision is designed, imprinting learning is taken into account. The designed circuit is mainly composed of imprinting learning module, generalization and differentiation learning module, multimodal learning module, behavioral decision module, graded response and feedback module. Behavioral signals are converted into high-intensity and low-intensity learning signals by imprinting learning module, which are output to multimodal learning module for multiple processes. The generalization and differentiation learning module is designed to better analyze the learning signals. Multiple factors are processed by behavior decision module, different behaviors are output based on GFMM neural network. The feasibility of the circuit is verified by PSpice, which provides a reference for biomimetic robots in learning, decision-making, and industrial firefighting.