Design of sliding mode controller for servo feed system based on generalized extended state observer with...

Area of Science:

• Control Engineering
• Robotics
• Mechatronics

Background:

• High-precision servo feed systems face performance degradation due to nonlinear friction, system uncertainty, and external disturbances.
• Achieving superior tracking accuracy is critical for advanced manufacturing and scientific instrumentation.

Purpose of the Study:

• To design a robust sliding mode controller for a two-axis differential micro-feed system (TDMS).
• To enhance tracking accuracy by mitigating nonlinear friction and system uncertainties.

Main Methods:

• A flexible two-mass drive model (FTMDM) was established for the TDMS.
• A generalized extended state observer (GESO) was developed to estimate system interferences.
• A sliding mode controller (SMC) based on GESO was designed and stability proven via Lyapunov theory.
• A double critic deep deterministic policy gradient (DCDDPG) algorithm was employed for dynamic parameter optimization of GESO.

Main Results:

• GESO demonstrated bounded observation errors and improved accuracy in estimating nonlinear friction.
• The proposed controller exhibited finite-time convergence of errors to zero.
• DCDDPG effectively reduced observation errors for step and sinusoidal signals.
• Experimental results confirmed enhanced position tracking performance on the TDMS.

Conclusions:

• The integrated GESO and DCDDPG approach significantly improves the tracking accuracy of high-precision servo feed systems.
• This control strategy offers a robust solution for complex dynamic systems affected by nonlinearities and uncertainties.