透過您的圖書館登入
IP:3.19.29.89
  • 學位論文

仿Type-2模糊控制器之應用於電動獨輪車

Application of Imitate Type-2 Fuzzy Controller for

指導教授 : 邱智煇

摘要


在本論文中,我們使用區間第二型模糊做仿製並根據第一代之電動獨輪車,變更了部分硬體加快其反應速度,且加強電動獨輪車的平衡控制之目標。我們設計了一個自我平衡控制器,利用傳統倒單擺控制之概念,以車身傾斜之角度與角速度當作控制變數,主動控制馬達的出力達到獨輪車直立不倒之目的。本論文中,我們主要提出了四種控制演算法來分別做比較,其中包含純PID控制器、第一型模糊控制器、區間第二型模糊控制器及我們提出的仿區間第二型模糊控制器。首先利用純PID控制器做定位實驗後再加入第一型模糊控制器做比較。然而,此電動獨輪車存在著許多不確定之因素及雜訊,會影響系統動作,為了解決此問題,我們使用區間第二型模糊控制器,但其模糊理論運算過程卻較為複雜,因此我們提出仿區間第二型模糊,其歸屬函數型態為兩個高斯函數,並藉由李亞普諾夫定理(Lyapunov theory)推導其學習速率且以達誤差之收斂;因此於本系統中我們將使用仿區間第二型模糊控制器來當系統主控制器。最後,我們也將透過電動獨輪車平臺之模擬與實驗,分別比較控制器模擬與實驗結果,驗證了仿區間第二型模糊控制器不但在平衡與控制上能有優異的成效也能有不錯的運算速度。

並列摘要


In this thesis, we use interval type-2 fuzzy to imitate the electric unicycle based on the first generation prototype in our laboratory. To enhance the goal of the balance control of electric unicycle, we have changed a part of hardware to speed up response and designed a balance controller via the concept of control of traditional inverted pendulum. We control the electric unicycle by the angle and angular velocity of electric unicycle. In this thesis, we compared four control algorithms including proportional-integral-derivative (PID) controller, type-1 fuzzy controller and interval type-2 fuzzy controller and imitate interval type-2 fuzzy controller. First, we compared type-1 fuzzy controller after the PID controller by positioning experiments. However, there are many uncertain noise and factors, so that the system motion is affected in the electric unicycle. In order to solve these problems, we use interval type-2 fuzzy controller. But computational process of fuzzy theorem that are more complexity, we proposed an imitate interval type-2 fuzzy controller, where the membership function are two gaussian functions, and its learning rate and error convergence are verified by the Lyapunov theory. In addition, we choose the imitate interval type-2 fuzzy controller as the main controller in this system. Finally, by the simulation and experimentation of electric unicycle, we compare all of controllers, respectively. From the results, the imitate interval type-2 fuzzy controller can be demonstrated that it not only has better performance for control and balance, but also good computational speed.

參考文獻


[37]吳俊達撰「大型雙足機器人之設計與實現:使用智慧型控制器」中原大學電機工程研究所,碩士論文,2011年(邱智煇、涂世雄指導)。
[59]周明鴻撰「電動獨輪車載具之設計與實現」元智大學電機工程研究所,碩士論文,2011年(邱智煇指導)。
[5] Toyota i-REAL,
[7] Toyota Winglet,
[13] Hammacher Schlemmer unicycle,

延伸閱讀