精密單層雙軸氣浮運動平台之迫近與定位控制

Abstract

本論文主要提出「迫近與定位控制」的控制架構，此架構欲解決在「速度和位置切換控制」中，速度控制和位置控制切換上造成的諸多問題，如速度和位置控制器頻寬不一致造成運動有不必要的震動，以及兩控制間切換時機點難以調整的問題，並提升系統運動的流暢性，減少平台抵達目標位置的時間，亦可使得系統控制參數調整上自由度較大。 此系統是以線性馬達以及空氣軸承所組成的雙軸氣浮平台。將雙軸的系統解偶成單一輸入、單一輸出(SISO)系統，後採用閉迴路系統識別對不穩定的系統進行數學模型的識別，再以多輸入、多輸出(MIMO)系統識別方法，加以探討系統的特性以及補償單一輸入、單一輸出數學模型的不足。 控制架構上，主要以狀態估測器估測系統無法量測的狀態變數以利狀態迴授控制，使系統能夠穩定。並以積分器處理穩態誤差的問題，整個架構在鏈散射式描述(CSD)的運算環境下，透過強健及最佳化控制來計算出常數控制的參數，使整個系統的H-infinite norm限制在一定範圍內。為了同時兼具速度與精度的表現，利用速度控制使平台高速移動，接近目標位置時切換到位置控制做精準的定位控制，稱之為「速度和位置切換控制」。而本論文提出的「迫近定位控制」中，輸入包含參考位置及參考速度，同時控制平台的速度及精度達到理想的響應，最終透過實驗可以發現「迫近與定位控制」既使在最大速度比較小的情況下，仍可在較短的時間到達目標位置。

This thesis proposes a novel control framework, “Access and Positioning control”, which is designed to solve problems caused while switching between speed and position control. For example, since the bandwidth of speed and position controller are not consistent, switching between these controllers may induce unnecessary vibration. Moreover, for the best performance, the switching timing is hard to determine. “Access and Positioning control” focus on boosting the fluency of motion and shortening the time which it takes for the stage to arrive the demanded target. The proposed method also makes the control parameters adjust easily. The system is a dual-axes air-bearing motion stage system consists of linear motors. For simplicity, in this thesis the complex system is decoupled to three single input single output (SISO) systems. Since the system is unstable, the closed-loop system identification is required. To compensate the inadequate parts of mathematical models from SISO identification, the multiple input multiple output (MIMO) system identification is utilized after the controller is designed. In pure speed or position control, the Luenberger observer is installed to estimate the states, which is not able to be measured by sensor for the state feedback control to stabilize the system. In order to deal with the steady state error, an integrator control is implemented. The overall control structure is based on the Robust and Optimal control to calculate the constant control gains in CSD, restricting the H-infinite norm of the system in a specific range. Performances of both speed and position control are important, thus a simple control strategy which applies the speed control first and switch to position control when the stage is near the designed target. This method is named “Speed and Position Switching Control”. On the other hand, the “Access and Positioning Control” includes the reference speed and position for input signals together so switching between two control methods is not necessary. The performance of speed and accuracy can be controlled simultaneously. The experimental results confirm that for “Access and Positioning Control”, it takes lesser time to arrive the demanded target even the maximum speed is lower than the “Speed and Position Switching Control” since the fluency of motion is raised.