GNSS-aided installation error compensation for DVL/INS integrated navigation system using error-state Kalman filter

Abstract

The integrated navigation system (INS), which integrates the strap-down inertial navigation system (SINS) and the Doppler velocity log (DVL), is widely used in the application of autonomous underwater vehicle (AUV) navigation and positioning. However, due to the constraints imposed by the AUV’s size and system design, a coordinate system mismatch occurs between the DVL and SINS, which can be modeled as installation error (IE), including installation error angles and lever arm errors that significantly impact the system’s accuracy. To address these issues, we introduce an INS based on feedback error-state Kalman filter (ESKF) that accounts for DVL installation error (IE) and propose a DVL IE compensation method. This paper details the system design of the ESKF-based INS, including the coarse initial alignment of the moving base, fine alignment, mechanization, and data fusion. The proposed IE compensation method is designed to successively estimate and correct the DVL IE, utilizing the global navigation satellite system (GNSS) for observations, minimizing the error model output to estimate the IE value. Simulations and field experiments indicate that the cumulative accuracy has been improved by 50.3% after compensation. Finally, carried by a disc-shaped AUV, the ESKF-based INS with DVL IE compensation performs well in practical AUV navigation applications.