Neural-like growing networks the artificial intelligence basic structure

Abstract

The article considers a new type of neural networks created by analogy with the neural network of the human brain and the theory of artificial intelligence which is based on these kind of networks. In this theory allocated three main functional units of integrative activity of the “brain” of intelligent robots: sensory systems; modulatory systems; motor systems. Here we consider the human sensory visual system and its model in intelligent systems. In this article are considered hypothesis about transformation the recognizable images to the same size in the fovea and image recognition on a subconscious level. Hypothesis realized in a hardware neural sensory model of a human sensory organ of visual system. As well, one can find the results of recognition of damaged images. In conclusion considers the block diagram of intelligent computer (electronic brain), which is represented as homogeneous, multiply connected, multidimensional, associative, active, growing neural matrix environment.