Implementing Inductive Logic into Embedded Industrial Control System with Polynomial Ring on F2

Minzhong Luo, Shan Yu

Abstract: In many industrial applications, planning control strategies in a way to guarantee that the collective behaviors of the system satisfies a certain steady specification is crucial. Motivated by this problem, in this paper, we develop an inductive logic representation method with polynomial ring on ${\mathbb{F}_2}$ and automatic control framework for industrial equipments. The goal of the industrial equipments is to accomplish diverse complex tasks, captured by learned rules set with inductive logic formulas. Specifically, we introduce a novel inductive logic with anonymous predicates to deal with the variability of industrial systems, and representation with polynomial ring on ${\mathbb{F}_2}$ is designed to make inductive logic reasoning computable. An industrial control simulation dataset containing multiple tasks is also constructed to evaluate the applicability of the method. In experimental simulation and practicing based on single-chip microcomputer, we illustrate the feasibility and efficiency of our proposed method compared with baselines on different industrial controlling scenarios by both rich numerical results and theoretical analysis.