Please use this identifier to cite or link to this item: `http://hdl.handle.net/11455/6897`
 標題: 全方位行動機械手臂之點對點最佳組態規劃與控制Point-to-Point Optimal Configuration Planning and Control of an Omnidirectional Mobile Manipulator 作者: 姜禮賓Jiang, Li-Bin 關鍵字: mobile manipulator;行動機械手臂;omnidirectional mobile base;image recognition;evolutionary algorithm;point-to-point trajectory planning;全方位行動;影像識別;進化演算法;點對點軌跡規劃 出版社: 電機工程學系 摘要: 本論文旨在發展一可被行動機械手臂抓取的圓形物體之影像辨識方法、全方位移動車體的運動學控制與基於進化演算法的全方位行動機械手臂之點對點控制方法與技術。一影像辨識方法被提出來辨識圓形目標物體和決定物體到攝影機之間的距離。本文提出兩個運動學控制法則來自動駕駛全方位車體，以到達所希望的位置或追隨設定的軌跡。一進化演算法被應用來搜尋行動機械手臂從一點移動到另一點之最佳化的組態。手臂動作之點對點路徑規劃是由一個簡單的三階方程式來完成。本文實驗用的系統是由影像辨識系統、移動車體系統、機械手臂系統與一台工業級電腦所組成。電腦模擬以及實驗結果證實本文所提出之理論之有效性。This thesis develops methodologies and techniques for image recognition of a circular object to be grasped by a mobile manipulator, kinematics control of an omnidirectional mobile base and EA-based point-to-point control for a mobile manipulator. An object recognition method is proposed to recognize the target spherical object and to determine the position of the object relative to the camera. Two kinematics PI control laws are presented to steer the omnidirectional mobile base to reach the desired position, and to drive the mobile base to exactly follow the desired trajectories. An EA-based task planning method is developed to find the near optimal configuration of the mobile manipulator moving form one point to another. A simple path planning for the point-to-point motion of the mobile manipulator is specified by a third order equation. The experimental system is composed of an object recognition subsystem, a mobile base subsystem, a manipulator subsystem and an industrial personal computer. Computer simulations and experimental results are included to verify the efficacy of the proposed methods. URI: http://hdl.handle.net/11455/6897 Appears in Collections: 電機工程學系所