A Fusion Algorithm of Underwater Dual-Sensor Gravimeter

Abstract

When strapdown gravimeter measures the underwater gravity field, the drift of inertial sensors, which is caused by thermal effects, degrades the performance. To solve this problem, a dual-sensor system for underwater gravity measurement is used. It is equipped with electrostatic suspended accelerometers (high-drift but high-resolution) and rotary modulation inertial measurement units (low-drift but low-resolution). For the dual-sensor gravimeter, we propose a new structure of the data processing algorithm. This Kalman-filter-based algorithm combines the advantage of two kinds of sensors to output low-drift and accurate gravity values. The data-fusion algorithm includes the drift estimation step and the Markov prediction step. The detailed structure and key issues are also described. Further, the performance of the algorithm is analyzed by simulation and real-world experiments. The result of real-world experiments shows that the accuracy of the gravity measurement in static conditions reaches 0.72mGal and the dynamic accuracy reaches about 1.96mGal.