音圈馬達防手震位置控制之FPGA晶片研製

Abstract

本論文使用反應快速的音圈馬達作為自動對焦系統的致動器，可應用於數位相機或數位攝影機，但由於使用者手部操作上的晃動造成系統產生摩擦力及傾斜角發生變動，常使得自動對焦系統無法達到理想的控制位置，因此在本研究裡，將導入干擾觀測器與反積分終結器配合比例積分控制器，來克服摩擦力及傾斜角變動所造成的擾動以實現音圈馬達防手震位置控制。 本論文以Altera公司所開發的場效型可規劃邏輯陣列(FPGA)晶片為基礎，整合數位邏輯晶片於單一顆FPGA晶片上，以實現音圈馬達防手震位置控制。在數位邏輯晶片中，以硬體描述語言(VHDL)實現數位轉類比控制程式、類比轉數位控制程式、數位濾波器、音圈馬達之速度PI控制器、位置P控制器、干擾觀測器、反積分終結器以及位置估測演算法等功能模組；在數值系統設計上使用24位元Q17格式的數值處理方式實現正規化，以解決數位邏輯晶片上浮點數運算的問題，且能夠提高音圈馬達控制精確度與晶片資料數值運算的解析度。此外，為了能有效降低自動對焦系統在數位相機裡的成本及體積，本研究採用磁性尺、小型的磁阻式感測器再配合軟體演算法取代使用昂貴的細分割晶片來獲取目前控制系統的位移跟速度資訊。 在實驗系統的建構方面，採用Altera公司Cyclone II系列的FPGA Development Board為音圈馬達的控制核心，並配合一套自行設計的外部電路，包含類比數位轉換、訊號驅動、訊號調整、訊號取樣的功能，來完成音圈馬達防手震位置控制系統的架構，並經由實驗數據的量測分析與控制參數的調整，來獲取系統較佳的控制效能表現。

In this thesis, the voice coil motor (VCM) with fast response time is used to be the actuator of the auto-focusing (AF) system applying to the digital camera or digital video camera. Because of the user’s handshaking, the AF apparatus induce the friction and inclination problem would cause incorrect position control. In order to overcome the effect of the friction and inclination, the disturbance observer and anti-windup with PI controller feedback system is applied to implement this anti-handshaking position control. The anti-handshaking position control of the VCM is realized on a chip based on FPGA (Field-Effective Programmable Gate Array) developed by Altera, and logic chips are integrated on a single FPGA chip. In digital logic chips, HDL (Hardware Description Language) is used to realize the function modules. For example, digital/analog conversion control, analog/digital conversion control, digital filter, proportional-integral (PI) controller for velocity loop, proportional (P) controller for position loop, disturbance observer, anti-windup controller, and position estimation algorithm. To realize normalization, 24 bits Q17 numerical format is used on numerical system design, and the problem of floating point number calculation on chip can be solved. Moreover, the precision of motor control and the resolution of chip data process can both be improved. Besides, the mini magnetic scale, mini magneto-resistive (MR) sensor and position estimation algorithm instead of using expensive interpolation chip are used for cost-down and mini-size. In the experiment system, Cyclone II FPGA Development Board is used to be the VCM control core, accompanying with a self-designed VCM Circuit Board consist of signal conversion, signal driving, and signal processing. The better system control performance can be achieved by measuring experiment data and modulating the control parameters.