透過您的圖書館登入
IP:216.73.216.225
  • 學位論文

無人載具操舵系統與即時控制器設計與實現

The Design and Implementation of Steering System and Real-time Controller for an Unmanned Vehicle

指導教授 : 王立昇

摘要


本研究主旨為建置具有轉向機構之無人載具系統,利用阿克曼轉向原理,使載具輪胎於轉向時能夠保持與速度同方向,產生滾動而不滑動的特性。此系統採用電腦視覺即時全域視覺定位(GVPS),用於判定載具位置與姿態、改良式A*路徑規劃使載具有效率的規劃避開障礙物的點到點路徑。 為了讓載具自主追蹤參考路徑,我們加入模糊控制器控制載具的角速度與轉向角,其中由載具與參考路徑點的距離誤差、視線角誤差和姿態角誤差組成三個輸入項以及一個載具角速度輸出項,讓載具能沿著路徑達成追蹤任務。 本研究將差速型轉向載具改為四連桿操舵轉向載具,搭配模糊控制器,及整合視覺定位和路徑規劃演算法,實驗結果顯示不論是直線、圓形路線亦或是規劃路徑的追蹤,無人載具都能順利完成追蹤路徑之任務。

並列摘要


The purpose of this research is to build an unmanned vehicle system that has the steering mechanism. Utilizing the Ackerman steering condition to keep the heading of the wheels being along the velocity when turning, the wheels roll without side sliding is guaranteed. To determine the position and attitude of the vehicle, the Global Vision Positioning System(GVPS) is adopted. Also, we apply modified the A*-algorithm to perform path planning from point to point without colliding with obstacles. To make the unmanned vehicle fulfill path-tracking mission automatically, we use fuzzy controller in the system, which can provide the appropriate angular velocity and steering angle for vehicle. In the controller, there are three inputs including the deviation of distance, the line of sight and the vehicle heading angle, and one output which is the angular velocity of vehicle. This research change the steering mechanism from speed-difference to 4-bar linkage and integrates fuzzy controller, GVPS and path planning algorithm. Experimental results show that for straight lines, circle, or other designed paths, the unmanned vehicle system can successfully track the desired path.

參考文獻


[1]蘇木春,張孝德,2012. ” 機器學習:類神經網路、模糊系統以及基因演算法則”,全華圖書。
[2]Li-Xin Wang, 1996. “A course in fuzzy systems and control”. Prentice Hall ISBN: 0135408822
[3]C.P. Papageorgiou, M. Oren and T. Poggio. 1998. “A general framework for object detection”. IEEE ISBN: 81-7319-221-9.
[4]Rainer Lienhart, Alexander Kuranov and Vadim Pisarevsky. 2002. “Empirical Analysis of Detection Cascades of Boosted Classifiers for Rapid Object Detection”. MRL Technical Report.
[5]Paul Viola and Michael Jones. 2001. “Rapid Object Detection using a Boosted Cascade of Simple Features”. IEEE. ISBN: 0-7695-1272-0.

延伸閱讀