Application of CUDA Computing Principles in Automatic Flight Control Simulation

The paper deals with aspects of mathematical modelling that is necessary for successful development of modelling and simulation technology. Several ongoing research efforts aim at improving automation of parallelisation for performance optimization of the Compute Unified Device Architecture (CUDA) software. An Application Processing Interface (API) allows the user to send commands to a graphics processing unit to draw objects on a display. The power of graphics processing unit is used by the simulation. Graphic Processing Unit (GPU) hardware designers have been developing unified processors which are increasingly resembling high-performance parallel computers. CUDA programming is an example of the well known single-programmultiple-data (SIMD) parallel programming method. This kind of programming is a popular method for massively parallel computing systems. When the mathematical models of aircraft control are appropriately designed, the simulation maybe implemented on the GPU. Mathematical models of the automated aircraft control are the systems, in which the regulation can simulate the program on the GPU. The design of mathematical model is analysed at the end of the paper.