An observation-based algorithm to identify the characteristics of a dynamic system (Q1883739)

The presented paper proposes a procedure for determining the nonlinear characteristics of dynamic system. It is based on the author's special orthogonal expansions and allows him to increase the number of identified parameters in the course of identification. Also it ensures the best use of the available measurement information. The procedure is illustrated by the determination of the aerodynamic characteristics of a model of a flight vehicle from the measurements of its motion along the aeroballistics route. The presented identification algorithm allows for attaining the maximal possible precision of approximation of the system characteristics. It has a wide convergence domain, which guarantees a solution even for rough a priori approximations of the estimated system parameters and characteristics. Also it is shown that the size of the convergence domain can be controlled by program or adaptive techniques or by varying the discretization of the continuation parameter within the interval of its measurement.