Hybrid intelligent system; Neural networks; Point localization

The article introduces a hybrid intelligent system for point localization in 3D Euclidean space. There are two models presented. The first one is based on neural networks and the second one represents a classical approach. The classical model calculates Euclidean distances between two points in the defined domain. As regards the experimental study, we proposed appropriate topologies of the systems that depend on the required accuracy. At first, we identified distances between a randomly generated point and a reference points in the defined domain. Then a neural network uses the obtained distances as its inputs to determine the actual position of the point in the domain space. The experimental study was repeated several times. All obtained results are mutually compared in the conclusion.