Memristor’s characteristics: From non-ideal to ideal

Abstract Memristor has been widely studied in the field of neuromorphic computing and is considered to be a strong candidate to break the von Neumann bottleneck. However, the non-ideal characteristics of memristor seriously limit its practical application. There are two sides to everything, and memristors are no exception. The non-ideal characteristics of memristors may become ideal in some applications. Genetic Algorithm (GA) is a method to search for the optimal solution by simulating the process of biological evolution. It is widely used in the fields of machine learning, combinatorial optimization, and signal processing. In this paper, we simulate the biological evolutionary behavior in GA by using the non-ideal characteristics of memristors, based on which we design peripheral circuits and path planning algorithms based on memristor networks. The experimental results show that the non-ideal characteristics of memristor can well simulate the biological evolution behavior in GA.