碩士 / 明志科技大學 / 機電工程研究所 / 97 / This study integrates global positioning system (GPS) and inertial navigation system (INS) by means of Kalman filtering (KF). Due to inadequacy related to KF-based GPS/INS integration, relatively poor positioning accuracy could be resulted during long GPS outages. To overcome this problem, covariance matrices that are applied in KF and associated with the system noise are tuned over time via neural network method. Therefore, the study highlights the use of neural network techniques to the adaptation of the initial statistical assumption of the KF caused by possible changes in sensor noise characteristics. The proposed system is implemented on a vehicle so that the desired positioning accuracy can be examined. It is found that the adaptive mechanism can help improve the positioning accuracy of the integrated GPS/INS system during long GPS outages.
Identifer | oai:union.ndltd.org:TW/096MIT00657015 |
Date | January 2009 |
Creators | Hung-Syi Lai, 賴鴻熙 |
Contributors | Jin-Wei Liang, Hung-Yi Chen, 梁晶煒, 陳宏毅 |
Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
Language | zh-TW |
Detected Language | English |
Type | 學位論文 ; thesis |
Format | 84 |
Page generated in 0.0099 seconds