Study on dual-layer MPC-SMC control strategy for piezoelectric platforms based on extended kalman filter

Area of Science:

• Robotics and Automation
• Control Systems Engineering
• Micro-manipulation Technology

Background:

• Piezoelectric stages in micromanipulation exhibit hysteresis nonlinearity.
• Image transmission delays further reduce positioning accuracy.
• Existing control methods struggle to compensate for these combined effects.

Purpose of the Study:

• To develop a novel dual-layer control strategy for precise visual servoing micromanipulation.
• To mitigate the impact of hysteresis nonlinearity and image transmission delay.
• To improve the overall tracking accuracy of micromanipulation systems.

Main Methods:

• Utilized image block matching and Gaussian kernel interpolation for high-precision displacement measurement.
• Integrated the Bouc-Wen (B-W) model with an Extended Kalman Filter (EKF) to model hysteresis nonlinearity.
• Designed a dual-layer control architecture combining Model Predictive Control (MPC) and Sliding Mode Control (SMC).

Main Results:

• The proposed dual-layer MPC-SMC strategy achieved a Root Mean Square Error (RMSE) of 0.0372 µm.
• This represents a significant improvement compared to individual SMC (0.1292 µm) and MPC (0.1366 µm) controllers.
• The strategy effectively compensated for hysteresis and transmission delays, enhancing tracking precision.

Conclusions:

• The B-W-EKF based dual-layer MPC-SMC control strategy is highly effective for visual servoing micromanipulation.
• This approach significantly improves positioning accuracy in the presence of nonlinearities and delays.
• The findings demonstrate a robust solution for high-precision micro-manipulation tasks.