Page 18 - Neural Network Modeling and Identification of Dynamical Systems

6 NEURAL NETWORK MODELING AND IDENTIFICATION OF DYNAMICAL SYSTEMS

problem is solved, as a result of which we get the of ensuring their resistance to changes in the
ANN representation for the aerodynamic drag properties of the object due to failures and dam-
coefficient of the maneuverable aircraft. ages; (3) an assessment of the nature of the im-
The Appendix presents the results of numer- pact of atmospheric turbulence on the controlled
ous computational experiments with adaptive system of the class under consideration; (4) an
systems of MRAC and MPC types with refer- assessment of the possibilities of adaptation of
ence to aircraft of various classes. These results the systems under consideration to uncertain-
provideanadequaterepresentationoftopics ties in the source data; and (5) assessment of the
such as (1) the effect of individual structural el- importance of adaptation mechanisms on the ex-
ements of these systems and the properties of ample of the problem of controlling the angular
these elements on the characteristics of the syn- motion of the aircraft.
thesized systems; (2) the capabilities of the sys-
tems under consideration from the point of view