| Summary: | 碩士 === 國立成功大學 === 測量及空間資訊學系碩博士班 === 100 === Currently, the Global Positioning System (GPS) is the most prevalent navigation method. However it suffers from signal blockage in places like urban canyons. The Inertial Navigation System (INS), on the other hand, provides position, orientation, and velocity information without external reference. But errors in INS accumulate over time, especially for low cost MEMS devices. This paper presents a post-processing method to improve GPS/INS navigation results by using auxiliary map information. When GPS temporally loses a signal, location can be estimated based on the map information providing driving direction and odometers, which can be used to compute velocities, as well as a digital elevation model (DEM) giving accurate height information of the estimated location. This paper presents a procedure to handle these three data sources and then integrate them into the Kalman filter algorithm. By approximating locations during signal blockage GPS/INS navigation solutions are significantly improved.
|
|---|
