LSSVM-PID based ship course control

FU Bin， LUO Wei-lin
(School of Mechanical Engineering and Automation， Fuzhou University， Fuzhou 350108， China）

Abstract： In order to solve the problems of the ship model identification in ship course control，the least squares vector machine（LSSVM） was applied to identification for ship heading model，and an inverse control system for ship course based on combined LSSVM-PID was investigated. Making full use of nonlinear mapping and parallel processing of the least squares vector machine，an inverse model of the ship course system was established，and this model was used as a feed-forward controller to design direct inverse control. Moreover，to improving the accuracy and anti-jamming ability of the control system，a PID control loop was added. The results indicate that LSSVM enjoys a sound nonlinearity and identification accuracy，and the compound control system has better dynamic tracking performance and anti-interference ability than the PID control system.
Key words： ship course control； least squares support vector machine（LSSVM）； inverse model control； proportion integral derivafive（PID） control