Combining Extended Kalman Filtering and Heuristics for Automotive Tracking

碩士 === 國立臺灣海洋大學 === 電機工程學系 === 98 === Nowadays, the prevailing GPS technology has been widely used in a variety of fields such as electrical consumer products, telematics and military implementation. The modern GPS system is capable of receiving signals from at least three satellites simultaneously...

Full description

Bibliographic Details
Main Author: 郭育昕
Other Authors: Mu-Der Jeng
Format: Others
Language:zh-TW
Published: 2010
Online Access:http://ndltd.ncl.edu.tw/handle/23222966119491832929
Description
Summary:碩士 === 國立臺灣海洋大學 === 電機工程學系 === 98 === Nowadays, the prevailing GPS technology has been widely used in a variety of fields such as electrical consumer products, telematics and military implementation. The modern GPS system is capable of receiving signals from at least three satellites simultaneously and regardless of the receiver’s location, a fast and accurate position determination of the receiver can be achieved with ease. However, GPS signal is easily affected by external factors such as weather and terrain. The effect can be critical especially in urban area, tunnel, underground car park and the confined space within a factory, where these adverse environments can result in a poor GPS signal or sometime no signal at all. Therefore, a mean to provide uninterrupted positioning services has been a crucial issue when it comes to navigation technology. This thesis aims at the study of enabling the navigation system to operate normally under any circumstances and providing reliable information on position. In this study, a microcontroller was connected to the OBDII interface scanner to gain more related data such as vehicle speed and etc. from the vehicle’s computer. The system also detects the voltage change as a result of the angular rotation of the gyroscope device and simultaneously receiving the GPS signals. The information were then be processed by the Extended Kalman Filter techniques to calibrate the error of the gyroscope. In time of the absence of signals, a scheme of positioning estimation with the aid of empirical Dead Reckoning (DR) system can be implemented to facilitate a continuous, uninterrupted positioning service.