| Stand-alone global positioning system (GPS) provides accurate long term navigation information but requires direct line of sight to at least four satellites, which are not always available. In contrast to GPS, inertial navigation system (INS) has good short term accuracy and operates autonomously, obviating the need for any external signal for the calculation of the navigation solution. Capitalizing on the complementary characteristics of these two systems, their synergistic integration promises to overcome the drawbacks of each other and provides more accurate navigation solution than either of them could achieve alone.;This research focuses on the implementation and verification of a tightly coupled INS/GPS integration architecture. In this architecture, the difference of the GPS computed pseudorange and pseudorange rate measurements and INS predicted estimates of pseudorange and pseudorange rate are input to a Kalman filter which uses them to estimate the errors in INS output. The INS output is corrected for these errors, providing an enhanced integrated navigation solution. The tightly coupled integration architecture has several advantages over the loosely coupled integration architecture. Tightly coupled integration eliminates the major restrictive requirement of loosely coupled integration which calls for at least 4 satellites to be visible for computation of a GPS navigation solution for INS aiding. Tightly coupled integration can provide GPS aiding even when the number of visible satellites is 3 or lower, thereby improving the operation of the navigation system in degraded GPS environments by providing continuous INS aiding even during limited GPS satellite availability. Tightly coupled integration also eliminates the problem of correlated measurements which arises due to cascaded Kalman filtering in loosely coupled approach. For the same hardware, tightly coupled integration almost always performs better than loosely coupled integration in terms of accuracy as well as robustness.;The tightly coupled algorithm was successfully implemented using MATLAB and tested for a tactical grade inertial measurement unit (IMU). Its efficacy was checked on three real-life trajectories by limiting the availability of satellites through artificial blockage of GPS signals. Several of these blockages were introduced and each blockage was repeated for 3, 2, 1 and 0 visible satellite cases. Duration of these GPS outages was 60 seconds which were introduced in various phases of a road trip to assess the performance of the algorithm. The results show adequate performance and verify the suitability of a satisfactory navigation solution during limited availability of satellites in most of the phases of a typical road trip.;Keywords. Inertial Navigation System, INS, Global Positioning System, GPS, Tightly Coupled Integration, Loosely Coupled Integration, Kalman filter.;With the developments in the field of inertial navigation and global navigation satellite system technologies, several integration schemes of INS and GPS have emerged with variety of implementations which include loosely coupled INS/GPS integration, tightly coupled INS/GPS integration and ultra-tight INS/GPS integration. Optimal estimation techniques, predominantly based on Kalman filter (KF), are usually employed to fuse the GPS and INS measurements, providing reliable and robust navigation solution. |
