Multi-sensor optimal data fusion for INS/GPS/SAR integrated navigation system

Abstract

Decentralized fusion and global optimal fusion INS/GPS/SAR integrated navigation system represents the trend of next generation navigation systems with the high performance of independence, high precision and reliability. This paper presents a new multi-sensor data fusion methodology for INS/GPS/SAR integrated navigation systems. This methodology combines local decentralized fusion with global optimal fusion to enhance the accuracy and reliability of integrated navigation systems. A decentralized estimation fusion method is established for individual integrations of GPS and SAR into INS to obtain the local optimal state estimations in a parallel manner. A global optimal estimation fusion theory is studied to fuse the local optimal estimations for generating the global optimal state estimation of INS/GPS/SAR integrated navigation systems. The global data fusion features a method of variance upper finiteness and a method of variance upper bound to ensure that the global optimal state estimation can be achieved under a general condition. Experimental results demonstrate that INS/GPS/SAR integrated navigation systems achieved by using the proposed methodology have a better performance than INS/GPS integrated systems.

Random weighting method for estimation of error characteristics in SINS/GPS/SAR integrated navigation system

“…GPS (Global Positioning System) is commonly used as an aiding sensor in SINS. However, GPS has the difficulty in providing continuous localization since the satellite signal may be lost and corrupted due to high buildings, tunnels and mountains, multi-path reflections and bad weather conditions [1][2][3]. It also has a low sampling rate.…”

“…It represents the trend of next-generation navigation systems with high independence, precision and reliability. However, the precision of SINS/GPS/SAR integrated navigation system exhibits statistical characteristics in terms of system error [1,4]. Therefore, statistical estimation of system error characteristics is an absolute necessity to enhance the positioning accuracy of integrated navigation system.…”

Cloud-aided SDR solution for lane-specific vehicle positioning via Local Interference Compensation

“…Antenna arrays, for example, would be an obvious possibility for compensating multipath effects, but require additional hardware and would increase the costs of any upcoming system [12]. The same holds true for Pseudolites Navigation [13], Multisensor Integrated Navigation [14], Antenna Design Techniques [15] or Autoregressive Modeling of INS Sensor Errors [16].…”

Fuzzy Kalman Filter Validation Using the Local Statistical Approach

“…By detecting local Kalman filters that perform estimation using an erroneous model it becomes possible to exclude them from the local estimates fusion procedure. Thus, the precision of the aggregate state estimate is assured, and this is particularly significant for the reliability of sensor networks and distributed autonomous navigation systems [24][25][26][27].…”