An Integrated Navigation Algorithm with Autoregressive Model for Land Vehicle Applications

• Main Authors: I-Feng Chien, 簡義峰
• Other Authors: Dah-Jing Jwo
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/52513649166754482837

碩士 === 國立臺灣海洋大學 === 通訊與導航工程系 === 98 === The advantage provided by Global Positioning System (GPS) is robust positioning. However it relies on the good GPS signal. For providing better positioning when the signal has been jammed, the integrated GPS/INS system is to be used. The advantage of inertial navigation system (INS) can provid continuous signal, while the signal of INS from a land vehicle is easy to be jammed. The jamming comprises the systematic and random components. This paper presents the systematic errors (deterministic) that can be estimated by calibration model, and the random errors can be estimated by an AR(autoregressive) processes. Many papers have presented the complementary advantages of GPS and INS. Loosely coupled is called when two systems are fixed each of position after two systems have been positioning; Tightly coupled is called when two systems are fixed at each position before two systems have been positioning; Ultra-tightly coupled is called when the GPS receiver have been tracking the signal, then INS signal is used to fix the GPS receiver signal. Autoregressive estimates in this paper specifically INS's bias. For autoregressive time series can be used to estimate the previous state the next state, this way to a more accurate approximation of actual bias in order to achieve the integrated GPS/INS system effects of elevated position. Keywords: Autoregressive process, Ultra-tightly coupled, GPS/INS