Modeling for helicopter based on support vector machine

Shu-zhou WANG,Ye SAN
DOI: https://doi.org/10.13196/j.cims.2008.03.56.wangshzh.008
2008-01-01
Abstract:To construct simulation model of a helicopter with high precision, the Support Vector Machine (SVM) method was introduced to the field of intelligent modeling of a helicopter. Based on pretreatment of practical flight data, such as eliminating outliers, filtering of high frequency noise, and differential smooth, the rotator speed model for landing process of a helicopter with rotator self-rotating was built up. Compared to the neural network model, the SVM simulation model of the helicopter was characterized by its simple structure, fast convergence speed and high generalization ability. It was revealed by theoretic analysis and simulation result that simulation model of a helicopter built by SVM method was feasible.
What problem does this paper attempt to address?