The main purpose of this project, funded by the U.S. Army Environmental Center, was to use an Airborne Ground Penetrating Radar (AGPR) system to locate and identify buried unexploded ordnance (UXO) at military sites. This project represents one component of the UXO Clearance Technology Program, an overall U.S. Government program, to clear former and present military ordnance ranges of all unexploded ordnance, and other buried devices that pose a threat to the public. As part of this government effort, Battelle, The Ohio State University and other contractors and government agencies have developed techniques that can rapidly and reliably determine the boundaries of contaminated sites, and the concentration of buried unexploded ordnance at those sites.

The part of the total system that was designed and built at The Ohio State University Center for Mapping (CFM) provides navigation (orientation) and positioning information to the system. Proper focusing of the AGPR measurements for Synthetic Aperture Radar (SAR) processing require high accuracy and high rate positioning information of the moving platform (helicopter). For this purpose the CFM has developed a high accuracy/rate airborne GPS/INS positioning system. This system integrates the Global Positioning System (GPS) and an independent Inertial Navigation System (INS). The integration of GPS and INS provides a survey system that exhibits superior performance in comparison with either GPS or INS stand-alone. Implementation of the On-The-Fly (OTF) ambiguity resolution technique enables precise cm-level positioning with initialization times of only 5 to 10 s, with as few as five satellites in view. This is essential to the airborne applications, especially when the moving platform is a helicopter, since tracking of all the satellites in view may be hindered by the interference of the satellite signals with the rotating blades. Combination of a high accuracy (cm-level) GPS positioning information with the inertial navigation measurements, provides a three-dimensional high-rate positioning between the 1 s (1 Hz) GPS position update, and during periods when GPS positions are not available due to aircraft banking, or due to interference with the rotating blades. CFM's GPS/INS system implements a loose, closed-loop GPS/INS integration. In the closed-loop configuration, that implies the feedback error calibration technique in the INS position computation loop, the INS position errors are limited to cm-level when the GPS is available at a 1 s rate. This greatly reduces the errors in the system model (a linearized form of the nonlinear navigation equation), and thus, increases the estimation accuracy of the optimal estimator implemented in the GPS/INS system. The CFM's system is capable of providing highly accurate position data in real-time or in post-processing.

The system was demonstrated to the Government during November 1995. The assets of the U.S. Army Aviation Technical Test Center (ATTC) at Fort Rucker, near Dothan, Alabama, were used in the demonstration. The Airborne GPR was tested at Fort Rucker and demonstrated at Tyndall AFB. The high performance of the integrated GPS/INS system, using actual flight data, was determined from the analysis of the differences between the predicted INS positions and the estimated GPS positions. The results show that they are at 0.01 m (1 s) level for all analyzed datasets, even when the INS was operating in a degraded NAV mode. Since the GPS positioning accuracy is at 0.01-0.03 m (1 s) level, the accuracy implied by the GPS/INS is at the order of 0.01-0.04 m (1 s) level for the times when cm-level GPS positions are available. However, the loose GPS/INS integration has a disadvantage of performing the positioning on the basis of INS measurements alone, when the number of tracked GPS satellites drops below four. This occurs when the helicopter maneuvers. The only solution to this problem is to implement tight GPS/INS integration, which allows use of GPS measurements from less than four satellites. Furthermore, tight GPS/INS integration provides faster and more reliable OTF ambiguity resolution after short periods of complete loss-of-lock to the GPS satellites. This approach is currently been implemented in the CFM's Airborne Integrated Mapping System (AIMS).