新型光學影像穩定系統開發與數位邏輯晶片實現

Abstract

本論文提出一應用於手機相機之光學影像穩定器(OIS)的動態建模、驅動系統設計與實驗驗證，此光學影像穩定器技術將改善手機相機在拍攝時，因手部晃動所造成影像模糊的問題，光學影像穩定器技術為在拍攝時，藉由微型制動器使鏡頭移動進行補償以提升拍攝影像之清晰銳利，使拍攝時得到更清晰的影像。目前多數手機相機之光學防手震(OIS)系統之微制動器均採用音圈馬達技術(VCM)。本設計使用兩軸平移式機構來進行鏡頭內光學模組及感光元件之位置補償，此機構由霍爾感測元件量測位移量，音圈馬達致動器來控制移動位置，達到拍照時的防手震效果。整個研究可以分成三大部份：系統鑑別與分析；相位超前－落後控制器設計；模擬與實作。第一部份將一光學防手震系統作為受控目標進行實驗，以頻率響應分析儀器測量測波德圖與運動方程式分析系統，建立兩軸(ｘ軸/ｙ軸)致動機構之光學影像穩定器。第二部份進行光學防手震系統之相位超前控制系統設計。第三部份為控制系統的模擬與實驗驗證，經由電腦軟體模擬與實驗來驗證控制器之功效後，再經由實際的電路來建立系統，再經由實際的電路來建立系統。此以現場可程式化閘陣列(FPGA)實現光學防手震系統之控制演算法，控制具有 光學影像穩定器，達到0.02秒內的安定時間，優於以往研究提出的0.1秒安定時間，穩態誤差小於2%，且在1~20Hz持續震動下鏡片震動訊號可降低至2.5微米。

This study proposes design, dynamic modeling and drive system of a two degree-of-freedom (DOF) translation optical image stabilizer (OIS) in the camera embedded in the mobile devices. This OIS differs from the previous designs since it stabilizes the shift between lens and image sensor by a two-DOF translational mechanism that is actuated by voice coil motors (VCM). It can provide a solution for a fuzzy image due to the hand-shaking. Such category of technology will enable cameras-phones images reaching more clear effect by using a micro-actuator to move the lens for compensation when taking photos. Currently, the micro-actuators of the OIS system used in mobile phone cameras are mainly based on the VCM technology. In this study, the control design uses the two-axis translational OIS, Hall sensors measure the displacement, and also uses a VCM micro-actuator to control the translational position of the lens. The work can be divided into three parts: the analysis of an OIS system and building x/y axis dynamic equation of OIS, the phase lead-lag controller design, and the experimental verification. In the first part, an OIS system is taken as a controlled target for the experiment, and the frequency response analyzer is used to measure the Bode plot and the analysis system of the equations of motion. Further, the second part is to conduct the phase lead-lag controller design of the OIS system. Lastly, the third part is the simulation of the controller and the experimental verification. After verifying the effectiveness of the controller through the Matlab simulation and the experiment, the system can be built with the actual circuit. In others words, this whole proposal can divide into three parts: (1) standardized system identification for OIS system; (2)design, simulation and validation of the Lead-Integral controller for varying parameters of OIS modules; (3)the digital control system development for OIS by Field-Programmable Gate Array (FPGA) evaluation and Matlab Simulink to modify parameters. The control ability of the OIS system which have moving range realized by FPGA can make the settling time be less than 0.02 second, which is better than the settling time around 0.1 seconds in previous studies, with a steady-state error less than 2%. Moreover, the OIS system can reduce the amount of movement to within 2.5μm averagely at 1-20Hz.