A New Coupled Method of SINS/DVL Integrated Navigation Based on Improved Dual Adaptive Factors

Abstract

The integrated navigation of strap-down inertial navigation system (SINS) and Doppler velocity logs (DVL) has the common application in the positioning of autonomous underwater vehicle (AUV). However, DVL may be interrupted for a short time when part of the beam cannot receive the reflected signal and DVL measurement information is easily affected by underwater complex environment and contains outliers. To solve the above problems, a Doppler shifts aided coupled method of SINS/DVL integrated navigation with pare of DVL beam measurements outages based on dual adaptive factors is proposed. In this paper, a tightly coupled approach for SINS/DVL based on Doppler shifts is proposed to solve the short-term failure problem caused by the lack of part measurement in DVL. Then a chi-square detection aided dual factors adaptive filter is used to suppress the outliers. The dual factors are used to adjust the influence of imprecise information of dynamic model and observation model error. The chi-square detection is used to judge the outliers of measurements and the result of judgment is regarded as the precondition of factor selection. In order to verify the robustness of the proposed algorithm to outliers, simulation and Yangtze River experiments are carried out. The results of simulation and river experiments indicate that the tightly coupled approach obtain higher accuracy compared by the loosely coupled model and the proposed adaptive filter can effectively suppress the outliers. The method proposed in this paper can achieve greater robustness in AUV underwater complex environment.