Please use this identifier to cite or link to this item:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/68485
Title: | 基於多模態融合之跟隨機器人 A Human-Following Robot Based on a Multi-Modal Fusion Strategy |
Authors: | Leong-Kian Tee 鄭良健 |
Advisor: | 黃漢邦 |
Keyword: | 多模態融合辨識,藍牙,模糊積分,跟隨機器人, Multimodal Recognition System,Bluetooth,Fuzzy Integral,Following Robot, |
Publication Year : | 2017 |
Degree: | 碩士 |
Abstract: | 隨著科技的進步,機器人領域的蓬勃發展,我們可以察覺到機器人已經慢慢的融入在這個社會上。其中最引人關注的莫過於服務型機器人,主要是因為與工業下的機器人相比,服務型機器人在與人互動這一方面考慮的因素要比工業下的機器人來得多。此外,人們對於服務型機器人更具有好奇心。為了達到機器人與人類之間的自然互動與和諧的共存,人機互動便是此發展重點。本論文致力於結合多模態融合辨識、手機藍牙以及雷射點所提供的距離資訊,來發展一個跟隨機器人,並達到當跟丟目標者時,服務型機器人能靠距離資訊去尋找回目標者,並從新進行辨識及跟隨。 With the advance of science and technology, robotics is becoming increasingly important. We can observe the integration of robots into our society. One of the most interesting aspects in the field of robotics is service robots. Compared with other robots used in the industry, service robots are required to be much more interactive with humans. People are therefore more intrigued by service robots. To achieve natural interaction between service robots and humans, human-computer interaction is one of the most important aspects of robotics. This thesis is focused on the development of a following robot based on a multimodal recognition system, distance information provided by Bluetooth from Android smartphones, and a laser rangefinder. When the following robot loses track of the target person, it has the ability to find the target person by using the distance information. After the target person is found by the following robot, the following robot can recognize the target person by using the multimodal recognition system. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/68485 |
DOI: | 10.6342/NTU201703926 |
Fulltext Rights: | 有償授權 |
Appears in Collections: | 機械工程學系 |
Files in This Item:
File | Size | Format | |
---|---|---|---|
ntu-106-1.pdf Restricted Access | 4.78 MB | Adobe PDF |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.