- 學位論文
移動型機器人於智慧型空間之導航與控制
Navigation and Control of a Mobile Robot in an Intelligent Space
摘要
現今研究中，架設在空間中攝影機的絕對位置多半設為已知，以利於移動型機器人在智慧型空間中導航與控制。為了解決在未知空間攝影機絕對位置下之機器人導航控制，本論文提出在未知空間攝影機絕對位置下以視覺為基礎之機器人控制系統。此系統運用機器人內建單眼攝影機來尋找空間中標有色碼之攝影機，並藉由所提出之選取法則切換空間中攝影機以控制機器人導航至目標點。本論文 提出兩種控制器以達到機器人定位任務，分別為固定增益控制器和模糊控制器。其中，固定增益控制器是以收斂角度及距離誤差來設計；模糊控制器則是以收斂距離、與影像水平軸角度誤差與機器人與目標物角度誤差所設計。本論文於室內空間中架設三台網路型攝影機以建構智慧型環境，並利用一搭載單眼網路視訊攝影機之自製移動式機器人驗證兩種控制器、攝影機選取法則及控制機器人在智慧型空間中導航至目標點。
並列摘要
For most of the existing studies, absolute positions of space cameras are mostly assumed to be known so as to facilitate navigation and control of a mobile robot in an intelligent space. For the purpose of navigating and controlling robots using video cameras with unknown positions, this thesis proposes a vision-based control system employing video cameras with unknown positions. The system uses a single on-board camera to search and identify the space cameras with color codes in the intelligent space and relies on the proposed selection rule to switch the space cameras in the feedback loop to control and navigate the robot to a target. This paper further designs a fixed gain controller and a fuzzy controller for the control purpose. Three IP cameras are installed in an indoor laboratory environment to build up the intelligent space. A custom wheeled mobile robot with a single on-board camera has been used to verify the effectiveness of the two controllers, the viability of the proposed selection rule, and the performance of controlling and navigating the robot to the target in an intelligent space.