A Direct Performance Comparison Between Loosely Coupled and Tightly Coupled GPS/INS Integration Techniques

Because of security concerns and systems integration complexities related to the implementation of a tightly coupled GPSllNS integration, a study was conducted to determine just how much of a performance advantage the tightly coupled technique offers over the less controversial loosely coupled (or cascaded) integration. The results of this study were: l Using four ideally positioned GPS satellites, loosely and tightly coupled performance were virtually identical. l Under degraded GPS satellite conditions, loosely coupled performance was significantly worse than tightly coupled performance. Further evaluation showed that this performance degradation was not only caused by the “cascaded filtering” inherent in the loosely coupled technique, but was also due to the fact that the GPS internal Kalman filter did not model the INS gyro and accelerometer errors that are modeled in the INS Kalman filter. When the GPS filter model was expanded to include these states, the resulting loosely coupled performance was comparable to the tight/y coupled performance under all conditions.