الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 231 |  |  |
| | | from | 231 | | |
Abstract
The integration of measurements from the Global Positioning System (GPS) with an Inertial Navigation System (INS) has been used in many applications of positioning and orientation. Over the last several years, a number of low cost Micro-Electro Mechanical Systems (MEMS)-based inertial sensors have been developed. Unfortunately, MEMS inertial sensor signals are contaminated by large random errors and the navigation solutions drift significantly over time. The overall objective of this dissertation is to investigate new methodologies to improve the performance of MEMS-based INS/GPS solution for hydrography and maritime navigation applications using two MEMS-based inertial units, namely ADIS16364 and DQI-100.
First, the effect of temperature point changes on low-cost, consumer-grade MEMS-based inertial sensor random error is examined. It is shown that the stochastic model parameters of a low-cost, consumer-grade MEMS-based inertial unit, namely the ADIS16364, are temperature dependent. Then, a novel, nonlinear stochastic model for inertial errors using a wavelet network (WN) is developed in this dissertation to replace the classical linear stochastic models, such as random walk (RW), Gauss-Markov (GM), and autoregressive (AR). It is shown that the first-order WN-based nonlinear stochastic model provides superior navigation results to the first-order GM and AR models. This new nonlinear model improves the performance of the ADIS16364 and DQI-100 units and introduces significant solution improvements when GPS outages exist, that range between 26% and 43% for 2D positions as compared to the standalone inertial navigation solution (prediction mode of the INS/GPS filter) when the current state-of-the-art linear GM-based is employed with the unscented Kalman filter (UKF).
A new frequency-domain dynamic response method is developed and applied to model INS/GPS integration and provide an accurate impulse-response-based INS-only navigation solution when GPS signals are denied (GPS outages). The transfer functions of the INS/GPS dynamic system are developed using the Least Squares Frequency Transform (LSFT). The discrete Inverse Least Squares Frequency Transform (ILSFT) of the transfer function is applied to estimate the impulse response of the INS/GPS system in the time domain. The new frequency-domain dynamic response bridging model improves the performance of the ADIS16364 and DQI-100 units and introduces significant solution improvements when GPS outages exist, that range between 22% and 26% for 2D positions as compared to the current state-of-art neural network bridging model. This new bridging model also introduces superior solution improvements when GPS outages exist, that range between 52% and 61% for 2D positions as compared to the standalone inertial navigation solution (without neural network aiding). Finally, the application of the newly developed methodologies for multibeam echosounder hydrographic surveys and maritime navigation is investigated vis-à-vis the International Hydrographic Organization (IHO) and International Maritime Organization (IMO) standards.