在本論文中,我們提出了一個用影像為行動機器人建立階層式地圖以達到快速定位之目標的系統。上層的地圖為一用圖形(graph)來表式的拓樸式地圖,因為此類地圖較適用於大環境。在這個地圖中,每個節點(vertex)由可以快速辨識其在整個環境中的那個地方(Place)的影像資訊所組成,而邊(edge)代表這些地方(Place)的空間關係。在此之後,我們會在每個節點之內建立一個以座標表示的地圖,其建立方式為影像資訊和由其他感應器得到的座標所成的對應。建完此地圖之後,可用即可利用兩個步驟達到快速定位的目標:首先,找出用來搜尋的影像落在那個地方(Place);再者,比對出此地方(Place)中與用來搜尋的影像最相像的影像,並傳回其所附之座標。
In our thesis, we propose a system to make the mobile robot building a vision-based hierarchical map and then quickly localizing itself by this map. In learning stage, each image The top-level map is a topological map which is represented as the graph because of the fitness of the large-scale environments. In this map, the vertices comprise some visual information for efficiently identifying the places of the robot's environment and the edges denote the spatial relation between these places. Then, we construct a small metric map attached each node of the topological map. After this two steps, the robot can localize itself by this hierarchical map effectively and robustly.