Vision based navigation system for autonomous proximity operations: an experimental and analytical study

Show full item record

Title: Vision based navigation system for autonomous proximity operations: an experimental and analytical study
Author: Du, Ju-Young
Abstract: This dissertation presents an experimental and analytical study of the Vision Based Navigation system (VisNav ) . VisNav is a novel intelligent optical sensor system invented by Texas A &M University recently for autonomous proximity operations . This dissertation is focused on system calibration techniques and navigation algorithms . This dissertation is composed of four parts . First , the fundamental hardware and software design configuration of the VisNav system is introduced . Second , system calibration techniques are discussed that should enable an accurate VisNav system application , as well as characterization of errors . Third , a new six degree -of -freedom navigation algorithm based on the Gaussian Least Squares Differential Correction is presented that provides a geometrical best position and attitude estimates through batch iterations . Finally , a dynamic state estimation algorithm utilizing the Extended Kalman Filter (EKF ) is developed that recursively estimates position , attitude , linear velocities , and angular rates . Moreover , an approach for integration of VisNav measurements with those made by an Inertial Measuring Unit (IMU ) is derived . This novel VisNav /IMU integration technique is shown to significantly improve the navigation accuracy and guarantee the robustness of the navigation system in the event of occasional dropout of VisNav data .
URI: http : / /hdl .handle .net /1969 .1 /1325
Date: 2005-02-17

Citation

Vision based navigation system for autonomous proximity operations: an experimental and analytical study. Available electronically from http : / /hdl .handle .net /1969 .1 /1325 .

Files in this item

Files Size Format View
etd-tamu-2004C-2-AERO-Du.pdf 1.002Mb application/pdf View/Open

This item appears in the following Collection(s)

Show full item record

Search DSpace

Advanced Search

Browse