- 學位論文
以重力感測器於攜帶式電話之距離估測
Distance Estimation with Mobile Phone Using G-sensors
摘要
本論文提出以重力感測器於攜帶式電話之距離估測，基於配置重力感測裝置的攜帶式手機上，我們藉由重力感測器所取得的數據進行目標距離的估測。 在本文中，首先我們將會介紹系統的整體架構，並說明本研究主要設計的距離估測以及誤差修正機制兩部分的內容。本系統建置於 Android 嵌入式系統平台上，並利用 Android 平台豐富的開發資源與重力感測技術做結合。第二，在距離估測部分，我們以三角測量法做為距離估測的基礎，透過重力感測器取得基線長度與水平角之角度，計算出與目標物間之距離長度。第三，考慮量測基線過程中所產生的誤差，因此我們將藉由誤差修正機制的方式，在系統操作介面中引導準確的基線路徑，以取得正確的基線長度數據。透過誤差修正機制的方式降低數據取得所產生的誤差，以提高距離估測的準確率。 最後我們將以實驗模擬來驗證本系統的可用性。我們將以實體重力感測裝置取得數據，透過 MATLAB 平台進行數據演算，來驗證距離估測的準確度。並於Android 的模擬環境下展示系統操作介面。 在本論文中，我們提出以下幾項貢獻： - 發展性：基於建構於 Android 平台下，系統有著開放性、豐富的開發資源以及硬體支援等特性，以提供本系統未來拓展開發的空間。 - 便攜性：只需要配置有重力感測裝置的攜帶式手機即可使用，屏除傳統距離量測所需的繁雜工具。 - 簡便性：在量測的過程中，只需幾項簡單的步驟即可完成量測過程，降低距離量測過程的困難度。 - 應用性：基於量測出與目標物距離之功能，將可用於各種需要測距的場合上，像是車載裝置、機器人避障、室內定位等各種應用上。 - 低成本：我們選擇使用率普及且價格便宜的攜帶式手機做為我們系統開發的裝置，相較於專業的測距儀器可節省額外的成本支出。
並列摘要
In this thesis, we propose a study on distance estimation with mobile phone using G-sensors. We get the acceleration data to calculate distance with target through mobile phone equipped with G-sensors. In the thesis, we will present system architecture and two designs focuses. The design focuses are distance estimation and error correction mechanism. At first, we build the system under Android development environment, and use Android development resources to combine with G-sensors. Thus, we use the abundant resources to develop the system. Second, we use G-sensor to get the acceleration and horizontal angle data to calculate distance with target. Based on trigonometric measurements method, we can calculate the distance from acceleration and horizontal angle data. Third, in order to avoid getting the error data, we use error correction mechanisms to reduce the error. In the user interface, it will lead to get the exact data and calculate distance with target. Through the method, we can reduce data error and increase the accuracy of distance estimation. At last, we will prove the feasibility of this system by simulation experiments. We use physical G-sensor device to get acceleration data and calculate object distance by using MATLAB. We will prove the accuracy of distance estimation and show the system interface under Android simulation environment. In this thesis, we propose following contributions: - Expansibility: Under Android development environment, the system has open, abundant resources and hardware assisted. It will provide the development space to the system. - Portability: If you want to measure distance with target, you just complete the action by mobile phone. It does not need many tools in the process of measurement. - Convenience: It just requires a few simple steps to complete distance measurement. It can reduce trouble process of measurement. - Adaptability: Based on distance measurement function, it can apply for any distance measurement cases (e.g., car device, obstacles avoidance robot, indoor location, etc.). - Lower cost: We use popular and cheap mobile phone to develop the system. It is cheaper than professional measuring equipment.