The application and improve of fuzzy clustering in classifying control targets

In this paper, a new clustering method is designed to solve the problem that control object model of a small aerial vehicle (SAV) will be changing accompany the changing magnetic field. The method is based on the establishment of fuzzy clustering approach. In this method, the variation of object model is used to establish the membership function to reflect the changing speed of the targets, and use experimental data to determine the parameters of the function which can ensure the authenticity of the membership function; using the robustness of the control system as the criteria for determining the cut-off matrix in order to ensure the reliability of the classification results. According to the test, the method can be classify the objects, according to prior experiment data, simulation data and the criteria of classification.