V. K. Abrosimov - Dr.Sc. (Eng.), Associate Professor, Moscow Aviation Institute (National Research University). E-mail: avk787@hotmail.com

New approach to the description of four-dimensional phase space (three coordinates and time) on the basis of neural networks with the ensemble organization oriented on the representation of movement and maneuvering of control objects in such space is provided. The idea consists in uniform filling of space with the neurons excited with a level of activity, appropriate probability or possibility of control object location in the space area simulated by neuron. The trajectory is simulated by excitation distribution on a neural network; and elements of excitation transmission matrix may be determined, stochastic or fuzzy. The effect of neural trajectories consists in the organization of parallel computings of trajectories for all control objects for group motion, determination of areas where objects can appear with the maximum reliability and possibilities of object group modeling simulation in case of principles of collective control implementation.