碩士 / 國立宜蘭大學 / 生物機電工程學系碩士班 / 102 / We developed a mobile robotic vehicle with the capability to patrol known indoor environment in order to resolve the blind spot issues of normal indoor video surveillance systems. This vehicle consists of three omni-wheels, a laptop, a BASIC stamp controller, ultrasonic sensors, and a laser rangefinder. The line features of the indoor environment were extracted and used to locate the vehicle position. The vehicle is designed to move around multiple rooms and corridors in the known indoor environment. It can also proceed to the assigned patrol points in sequence.
The experiments show that the laser rangefinder is able to measure line features under different measuring lengths and angles. The error rate is below 5% when we measure the flat wooden board with fixed angle and variable distances. The error rate is below 4% when measuring with fixed distance and variable angles. The experiments also show that the program can successfully reconstructs the hidden line features with the error rate of the length under 3％. The error rate using an encoder to measure the moving distance of the vehicle is below 6% within 0.5m of linear motion and is below 4% for rotary motion of the vehicle.
The experiments also show that the program is capable of locating the vehicle position in the maps. The distance error is less than 0.05m with a laser rangefinder moving around in a single room formed by rectangles. The finally developed vehicle is able to patrol in multiple rooms successfully. The vehicle can relocate itself and complete the patrol even when locating failure happened sometimes.
| Identifer | oai:union.ndltd.org:TW/102NIU00730009 |
| Date | January 2014 |
| Creators | Yi-Xuan Huang, 黃浥諠 |
| Contributors | Feng Ou-Yang, 歐陽鋒 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 169 |
Page generated in 0.002 seconds