Video Updates during GPS Dropouts using Navigation and Electro/Optic Sensor Integration Technology

The authors have developed Navigation and Electro/Optic Sensor Integration Technology (NEOSIT) that operates by using video navigation updates (VUPTs) to maintain inertial navigation solution accuracy during GPS dropout periods. The navigation updates are provided from object reference models downloaded pre-mission. These can include features such as buildings, road intersections, or bridges. To reduce computation, an inertial-aided model matching technique is used. The NEOSIT software includes highly efficient image processing algorithms that use the integrated navigation and optical sensor data to locate features captured within the digital image data. In the summer of 2001, the NEOSIT system performance was demonstrated using an integrated GPS/inertial/video system to collect imagery and navigation data over Tift County, Georgia. Object models were generated from this data set, with a differential GPS (DGPS) aided navigation solution. A second set of navigation data was then generated, without the use of DGPS aiding, to simulate a GPS dropout. Using the NEOSIT model detection software, a navigation solution was computed using inertial navigation data with image aiding. The difference between the DGPS-aided truth source and the image aided navigation solution demonstrated the ability of the NEOSIT system to navigate accurately in the absence of GPS aiding.