The design of self-healing memristive network circuit based on VTA DA neurons and its application

Abstract

This paper proposes a self-healing memristive network circuit, which simulates the resilience mechanism of VTA DA neurons. The proposed memristive network circuit mainly consists of four modules: input (synapse) module, damage detection module, threshold trigger module and negative feedback module. The input module is composed of a memristive synaptic circuit, which can react to a harmful initial input rather than a normal initial input. Apart from reacting to initial input, the input module also can receive a feedback signal released by the negative feedback module, and the process of receiving feedback signal is the key of realizing the self-healing function. The damage detection module can output a low level pulse that acts as an input signal of the threshold trigger module when the harmful initial input exists. The threshold trigger module will activate the negative feedback module when the trigger point which corresponds to the positive threshold voltage of memristor is reached. As the last module of the self-healing function, the negative feedback module can transmit the feedback signal, also be called a repairing signal, to the damaged memristor of the input module. The PSPICE simulation results indicate that the proposed memristive network circuit can realize the self-healing function which corresponds to the resilience mechanism of the VTA DA neurons. Meanwhile, when the proposed memristive network circuit has been applied to the damaged electronic machines or robots, the self-healing function of it can make the machines be reusable.