Application of Support Vector Machines to the Modeling and Control of a UAV Helicopter

This paper presents an application of Support Vector Machines (SVMs), based on statistical learning theory, to the dynamic modeling and automatic control of UAV helicopters. The ultimate goal is the development of an Artificial Intelligence (AI) module for the UAV system. The primary objective of the artificial intelligence unit is to train the UAV to operate along a desired path. Based on the data obtained during flight test experiments, an artificial intelligence model is trained. The process of training and predicting the path of the flight is a problem of estimation. Use of Support Vector Regression is shown as applied to the parameter estimation of a UAV helicopter. The developed model will be used in the artificial intelligence module that will be able to send out appropriate signals that enable the UAV to maintain the desired path while making minimal assumptions about the initial flight conditions.