本論文設計模糊控制器來控制小型人形機器人的手臂,讓視覺自主人形機器人可以完成「FIRA世界盃機器人足球賽:人形機器人組(HuroCup)」之籃球(Basketball)競賽項目,所提的方法可以加快與提高機器人投籃的完成速度與準確度。本論文主要有2個部分:(1)基因演算法(Genetic Algorithm)與(2)手臂控制器的設計與實現。在基因演算法方面,本論文使用基因演算法來模擬機器人投球動作,找尋影響投球距離重要的關係因素,確定以肩關節伺服機速度做為手臂控制器的控制輸出。在手臂控制器的設計與實現上,本論文設計一個兩輸入一輸出的模糊系統,兩個輸入分別為視覺系統所量測的距離以及伺服機驅動電壓,輸出為肩關節伺服機速度,此模糊系統會根據視覺系統所量測的距離以及目前伺服機驅動電壓輸入的大小決定出一個肩關節伺服機的速度。從實驗結果可知,所提的方法確實可以加快與提高機器人投籃的完成速度與投籃命中率。
In this thesis, an arm controller of the humanoid robot based on the fuzzy controller is designed to speed up the motion and improve the shot accuracy for the basketball category in Humanoid Robot Cup (HuroCup) of FIRA World Cup Robot Soccer. There are two main parts: (1) Genetic Algorithm (GA) and (2) arm controller. The shot motion is simulated in GA to find the key joint, shoulder joint, which influences the shoot distance the most. The rotation speed of the shoulder joint is used as the controller output. In the design and implementation of the arm controller, a two-input-and-one-output fuzzy controller is implemented. The inputs of the fuzzy system are the distance between the robot and the basket obtained by the image system and the input voltage of the shoulder joint. The output of the fuzzy system is the rotation speed of the shoulder joint. The proposed fuzzy system determines the rotation speed based on the distance and the input voltage. From the experiment results, we can see that the effectiveness of the proposed design method.
為了持續優化網站功能與使用者體驗,本網站將Cookies分析技術用於網站營運、分析和個人化服務之目的。
若您繼續瀏覽本網站,即表示您同意本網站使用Cookies。