碩士 / 國防大學理工學院 / 資訊工程碩士班 / 104 / With the lately popularity of mobile smart devices embedded with multi-axes sensors, many applications applied sensor fusion techniques of mobile smart devices to assist the user to perceive surrounding environment and further to provide the services what users need. The thesis proposes the sensor fusion algorithm to integrate the common image sensor and multi-axes sensors of mobile smart devices for indoor localization application. The proposed approach bases on spatial and geometric features of floor-plan building to evaluate the different sensor fusion algorithms to have adaptive threshold to filter preliminary noise and to find the most appropriate parameters for user’s behavior. The classified behavior includes standstill state, stand around state, moving forward state, moving around state, and lost focus state, etc. Based on inertial user’s behavior state to integrate indoor image sensor localization method, it provides stable and in-time indoor localization results, finally. In order to follow the future development of smart glasses, the study simulates the smart glass to put the smart device on the chest. The experimental results show that the proposed sensor fusion technologies can effectively improve the inconsistent sampling rate problems which produced by legacy sensor fusion algorithm. The results further applied augmented reality techniques to implement live-view indoor navigation systems. The proposed study can be used in smart firing, anti-terrorism, indoor warrior training, house navigation and internet of things applications.
| Identifer | oai:union.ndltd.org:TW/103CCIT0394003 |
| Date | January 2016 |
| Creators | Chang, Huai, 張懷 |
| Contributors | Tsai, Chung-Hsien, 蔡宗憲 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 53 |
Page generated in 0.0022 seconds