Aided Navigation: Theory and Applications for Sensors and Architectures

The coupling of GPS and INS is often characterized as loose, tight, or ultra-tight depending on the nature of the GPS input to the navigation filter. Thus, considering the challenge of precision navigation without GPS, this project will test and validate the coupling of sensor (EO, IR) information such as ego-motion and position updates similar to the coupling of a GPS and INS system. This project draws from two technical fields: image processing and filter design. Based on our related work, image processing in context of developing robust algorithms for position and ego-motion estimation, is for the most part complete. Thus, most of the Phase I effort will focus on the filter design aspect of the problem. The primary objectives of this proposal are to (1) develop models representative of various types of coupling and (2) predict system level performance for each of these models based on measures of performance such as effective IMU drift, and navigation accuracy with and without sensor aiding. The project team is led by Scientific Systems Company Inc. (SSCI) and includes the University of South Florida (USF) as its academic partner.