碩士 / 淡江大學 / 電機工程學系碩士班 / 103 / This paper proposes a VFH-based (vector field histogram based) hybrid navigation system. This system can execute global path planning and local path planning based on the rough map. The VFH-based hybrid navigation system can be described into two layers. The first layer is deliberative layer and the second layer is reactive layer. In the deliberative layer, the system automatically builds a grid map based on the rough map, which the robot has known. The concept which closing operation of image processing is used to modify the grid map to avoid the robot enters some dangerous areas. In the global path planning, the system uses an A* modified algorithm to execute the global path planning and extract the turning points on the path. In the reactive layer, the system uses local path planning algorithm to avoid some obstacles between two turning points. In this paper, the VFH algorithm is used to create a histogram via the laser range finder. A cost function is used to decide the best moving direction to avoid the obstacles. This paper also implements a virtual force field (VFF) algorithm to compare with the VFH algorithm. Form the results, we discusses the advantage and disadvantage between the two methods. Finally, this system makes the robot to avoid unexpected obstacles safely and move effectively in the home environment.
| Identifer | oai:union.ndltd.org:TW/103TKU05442006 |
| Date | January 2015 |
| Creators | Ming-Hua Hsieh, 謝明樺 |
| Contributors | Shih-An Li, 李世安 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 59 |
Page generated in 0.0125 seconds