Please use this identifier to cite or link to this item:
標題: 輪式行動機器人之影像定位與路徑追蹤
Vision-based Localization and Path Tracking of a Wheeled Mobile Robot
作者: 汪學進
Wang, Xue-Jin
關鍵字: Wheeled Mobile Robot;輪式行動機器人;triangulation method;extended kalman filter;polar space;backstepping;三角測量;增廣型卡爾曼濾波器;極座標;倒逆步
出版社: 電機工程學系
本論文旨在針對輪式行動機器人,發展以視覺為基礎之自我定位及發展一種以極座標空間表示之路徑追蹤控制法則。 根據三角測量方法,本文利用攝影機與三個指定的彩色地標可計算得到的機器人在一平坦地面環境之姿態。再者結合增廣型卡爾曼濾波器策略(extended Kalman filter)來融合視覺定位和里程計,可改善輪式行動機器人姿態估測之精準度。另外,為開發輪式機器人於極座標底下之路徑追蹤控制技術,倒逆步的方法以及李式(Lyapunov) 穩定理論被應用於發展出一新型穩定的路徑追蹤控制策略。電腦模擬和實驗結果顯示本文所提出的姿勢量測、追蹤和路徑追蹤控制的有效性。

This thesis develops methodologies and techniques for vision-based self-localization and polar-space path tracking of a wheeled mobile robot. Based on the triangulation method, posture information of the robot at a flat environment is determined using a CCD camera along with three given colorful landmarks. An extended Kalman filter (EKF) algorithm is employed to merge both the visual and odometric measurements for improving the pose estimation accuracy of the robot while traveling over its working space. To achieve path tracking in polar space, a new stable nonlinear tracking control law is developed using the backstepping approach and the Lyapunov stability theory. Several computer simulations and experiment results are provided to show the effectiveness of the proposed posture tracking and path tracking methods.
Appears in Collections:電機工程學系所

Show full item record

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.