Abstract: Compared with the surface of the water, it is more difficult to accurately estimate the E?tv?s effect in underwater, which affects the accuracy of gravity measurement. The traditional gravity matching algorithm does not consider the impact of E?tv?s estimation error on the performance of gravity aided navigation. According to investigating the degree of influence of different navigation parameters on E?tv?s estimation error, establishes inertial navigation system (INS), Doppler velocity log (DVL) and gravimeter underwater navigation framework based on unscented Kalman filtering (UKF). The performance of gravity aided navigation with or without E?tv?s estimation error is evaluated. Semi-physical simulation results show that the DVL underwater aided navigation method can restrain the accumulation of INS errors to a certain extent. The positioning performance of the INS/DVL/Gravimeter navigation mode is further improved. Even when the E?tv?s estimation error is considered, the horizontal radial positioning accuracy is still improved by 17.02% compared with only using DVL as the aided information.

Key words: Inertial navigation system, Gravity aided navigation, Kalman filter