利用非線性控制使四旋翼飛行器達成軌跡的追蹤

隨著無人飛行器逐漸應用廣泛，設計飛行器的控制系統也漸漸變成熱門的研究議題。本篇以設計四旋翼的控制器為主題，首先建立物理動態模型，根據此模型設計控制器，其主要是由兩個控制器組成: 基準控制器及適應控制器。以基準控制器為例，基準控制器是以步階回歸來設計，作為軌跡追蹤的主要控制器 ; 而適應控制器則是根據模型參考適應控制，用來增加系統的強健性，並提供更好的穩態追蹤效果。為了測試所設計的控制器，我們提供了模擬及實際四旋翼飛行器的室內飛行數據來佐證其可性度。在取得飛行器空間位置上，我們使用其自身感測器(加速度計、陀螺儀)來做換算，並使用立體攝影機(Kinect)來得到更完整的環境狀態。利用Kinect相機所提供的深度資訊，根據影像與感測器相對關係獲得自身三維空間位置以獲得四旋翼的空間位置資訊。

關鍵字

立體攝影機；三維座標重建；四旋翼飛行器；步階回歸；適應控制

並列摘要

As the unmanned aerial vehicles have been increasingly popular, designing a control scheme for unmanned aerial vehicle becomes a popular issues. In this thesis, the dynamic model of quadrotor is first derived. Based on the model, a controller which is a combination of baseline and adaptive controllers is presented for accomplishing the desired tasks. The baseline controller based on backstepping technique is the main controller for trajectory tracking; the adaptive controller is found to increase robustness to steady-state error using Model Reference Adaptive Controller (MRAC). Then, simulations and flight experiments are given for testing the robustness and validity of the controller. In experiments, the position of quadrotor is obtained by the external motion sensor and IMU including gyroscope and accelerometer. The external motion sensor is constructed by RGB-D sensor, Microsoft Kinect. Using the depth information from Kinect, the 3D position of quadrotor could be obtained based on the projection relation of camera.

並列關鍵字

RBG-D sensor ； three-dimension coordinate reconstruction ； quadrotor UAV ； backstepping ； adaptive control

參考文獻

[1: Islam et al. 2015]

S. Islam, P. X. Liu, A. El Saddik, “Robust Control of Four-Rotor Unmanned Aerial Vehicle with Disturbance Uncertainty,” IEEE Transactions on Industrial Electronics, vol. 62, no. 3, pp. 1563-1571, Mar. 2015.

[2: Nemra & Aouf 2010]

A. Nemra and N. Aouf, “Robust INS-GPS Sensor Fusion for UAV Localization Using SDRE Nonlinear Filtering,” IEEE Sensors Journal, vol. 10, no. 4, pp. 789-798, Apr. 2006.

[3: Muller et al. 2011]

國際替代計量

利用非線性控制使四旋翼飛行器達成軌跡的追蹤

全文下載

主題瀏覽