Comparison of Centralized and Decentralized Kalman Filter for SAR/TRN/GPS/INS Integration

This paper presents a performance comparison of centralized and decentralized integration of global positioning system (GPS), synthetic aperture radar (SAR) landmark detections and terrain referenced navigation (TRN) system into an inertial navigation system (INS). In recent years many publications have shown different decentralized integration schemes. Especially the federated no-reset filter has received considerable attention in literature. The main advantage of the decentralized filter architecture is improved fault detection and isolation (FDI) capability which is a very important feature in multi-sensor navigation systems. Improved fault detection capability of the federated no-reset filter is well known. However, this filter type combines sensor data in a suboptimal way. Generally a loss of accuracy compared to the centralized Kalman filter is observed. Earlier publications mentioned this loss of accuracy, but did not give the amount of accuracy loss for SAR/TRN/GPS/INS systems. This paper gives an insight into the relative accuracy loss of the federated no-reset filter compared to the centralized filter depending on different IMU types.