Friction modeling and compensation in motion control system using support vector regression (SVR)

A non-parametric-based technique using epsiv-insensitivity support vector regression (epsiv-SVR) is proposed and developed in this work for the identification of friction in a DC motor-driven motion control system. The effectiveness of the developed SVR-based friction model to compensate the frictional effects is evaluated experimentally on a rotary experimental set-up for both point-to-point (PTP) and tracking positioning control. The results obtained show SVR as a viable alternative to mathematical-based techniques in representing and compensating friction effects.