A Compensation Algorithm with Motion Constraint in DVL/SINS Tightly Coupled Positioning

Abstract Integration of strapdown inertial navigation system (SINS) and doppler velocity log (DVL) is usually applied in underwater applications, wherein DVL provides the three-dimensional velocity, and hence the accumulated error of SINS can be compensated. However, the DVL/SINS loosely coupled approach fails in the complex environments on the condition of fewer than three available beams. And the tightly coupled approach divergences with only one direction velocity information from the beam measurement. To solve the problems, a novel tightly coupled method is proposed in this paper, in which the state of UUV motion is considered, and a robust adaptive Kalman filter is utilized to dynamically estimate the observation noise. Experiment results indicate that the proposed method is efficient in UUV missions for beam limited environment.