Research On SINS/USBL/DVL Integrated Navigation System And High Performance Algorithm For Underwater Vehicles

With the development of underwater vehicles, underwater navigation system has much higher requirements in terms of accuracy, reliability and so on. Single navigation system has been difficult to meet the requirements, the integrated navigation technology which based on the multi-sensor information fusion becomes the development trend of the underwater navigation. The Integrated navigation system, which has synergistic beyond, complementary and redundant functions, can make full use of the navigation information of each subsystem, complement each other and improve the navigation performance of the whole system. Multi-sensor data fusion is the key to realizing integrated navigation, and the emergence of various filter algorithms provides a solid theoretical foundation and powerful mathematical tool. Therefore, the research of underwater vehicle integrated navigation system and high performance filter algorithm has been paid attention by many scholars in the fields of marine, navigation, guidance and control, information fusion and so on, and has become one of the hot research issues.Based on comprehensively analyzing the advantages and disadvantages of strapdown inertial navigation system (SINS), ultra-short baseline(USBL) and Doppler velocity log (DVL), the SINS/USBL/DVL integrated navigation system is designed in this paper. Then, a noval adaptive unscented Kalman filter(UKF) based on maximum likelihood principle and receding horizon estimation is put forward for overcoming the existed problem of UKF. Finally, the proposed adaptive UKF is applied to the designed SINS/USBL/DVL integrated navigation system and carried on the experiment research and analysis.The main research work and innovative contributions are as follows:1. The SINS update algorithm, USBL positioning principle and DVL velocity principle are studied. Then, SINS nonlinear error model is established for overcoming the existed shortcomings of linear error model. Finally, USBL positioning error and DVL velocity error are analyzed.2. Based on the research of integrated navigation mode, the SINS/USBL and SINS/DVL integrated navigation subsystems are designed and the nonlinear mathematical model of each subsystem is established. Then, the SINS/USBL/DVL integrated navigation system is designed by applying the federated filter technology.3. Aiming at the problem that the filter precision of UKF is not high or even divergence under the condition of unknown or inaccuracy noise prior statistical properties, an adaptive UKF based on maximum likelihood principle and receding horizon estimation is presented. The innovation sequences generated by UKF are applied to construct the likelihood function which includes noise statistical properties, and then the concept of receding horizon estimation is introduced to restrain the divergence of noise statistical estimation by online receding horizon optimization. Finally, Newton-Raphson algorithm is used to calculate the maximum likelihood estimation of noise statistics and the adaptive UKF with noise statistics estimator is obtained.4. The proposed adaptive UKF is applied to the designed SINS/USBL/DVL integrated navigation system, and verified by the simulation and actual shallow water experiment. The results demonstrate that the designed SINS/USBL/DVL integrated navigation system, which has high navigation accuracy, can meet the requirements of underwater vehicle accurate navigation. The proposed adaptive UKF can effectively restrain the influence of the unknown or inaccurate noise statistics on the positioning accuracy of integrated navigation system, thus improving effectively the underwater positioning accuracy of SINS/USBL/DVL integrated navigation system.The research of underwater vehicle navigation technology is a multi-disciplinary research topic and needs to do a lot of research work. The research work of this paper not only has important academic value for underwater navigation technology, but also makes some contributions to the fields of navigation, guidance, information fusion and so on.