Research On Sea Current And UUV Velocity Estimation Methods Based On Model-Aided Navigation

Sea current and velocity measurements from a Doppler velocity log (DVL) have an important impact on accuracy of traditional dead-reckoning by combining attitude information and DVL directly, when Unmanned Underwater Vehicle (UUV) is long cruising voyage. Vehicle velocity error will cause navigation error. Moreover, velocity measurements will be unavailable when distinct change in the bathymetry lead to DVL loss of bottom track,or DVL fails to work, then additional aiding is needed to estimate velocity. An accurate vehicle kinetic model can provide effective water-relative velocity, and then vehicle velocity relative to the bottom can be obtained by combining the sea current and the velocity calculated by kinetic vehicle model. And sea current estimation is of great importance when UUV can only obtain the water-relative velocity. Therefore, research on sea current and vehicle velocity estimation method based on kinetic model is carried out in this paper.The main contents in this paper are shown as following.Firstly, this paper applies the nonlinear wavelet threshold denoising method on sensor data denoising prior to being utilized. DVL error analysis is presented, and an adptive identification method is used to estimate the DVL alignment. The performance of this adptive identification method is evaluated by simulation experiment. Furthermore, the improved Sage-Husa adaptive Kalman filtering method is utilized to estimate vehicle velocity to improve DVL measurements accuracy.Secondly, considering the problem of asynchronous sensor registration when integrating measurements from different sensors, an improved interpolation method is applied on registration to the same sampling frequency. For sea current estimation when DVL is valid,a perturbation method is used for deriving the velocity error states equation, and model-aided DVL Kalman filter is designed. Simulation experiment using sea trial data evaluate the feasibility of the method.Thirdly, for vehicle velocity estimation when DVL is invalid, by analyzing the relationship between sea current, water-relative velocity and bottom-relative velocity, and considering vehicle velocity may change, a weighted index recursive least square estimation method based on kinetic vehicle is proposed to estimate vehicle velocity, and the sea current is also estimated. The feasibility is validated on experimental data collected by sea trial.At last,utilize experimental data collected by sea trial and the proposed method in this paper to estimate sea current and vehicle velocity. And apply the results on UUV navigation when DVL is failure. By comparison of simulation result and sea trial data, the feasibility of the proposed method is validated.