Nonsingular Fast Terminal Sliding Mode Control

At present, many researchers have put forward a lot of control strategies to improve negative problems of pneumatic cylinder servo systems, which is caused by nonlinearity of servo valves, various friction forces, air compressibility and so on [1, 2]. A neural network control scheme based on PID is introduced to ensure good tracking performance of a pneumatic X–Y table [3]. Methods in [4, 5] are presented to modulate pneumatic servo systems with variable payload. However, these control schemes require highly complete mathematical models which are hardly obtained because of strong nonlinearity, unmodeled dynamics and external disturbances. It is a key idea to estimate these uncertain factors effectively by using an ESO for a pneumatic rod cylinder servo system.