AUV Integrated Navigation Algorithm Study And The System Implementation Based On MOOS

Autonomous underwater vehicle （AUV） has been a subject of intense interest in the field of oceanic engineering in recent years. AUVs can be equipped with various sensor devices to collect data effectively, they become more and more widely used in the detection of underwater environments.The navigation technique is a key for the development of AUVs. Because of the GPS can not be used underwater, and the factors such as long operation time, requirement for high level of covertness and reliability, it is a great challenge to efficiently achieve the autonomous navigation for AUVs. In practice, a commonly-used way is the integrated navigation by using a variety of sensor data, it can reduce the error that would increase fast with time when using the traditional dead reckoning navigation system.This paper introduces the application of the algorithms based on Kalman Filter_KF for the AUV navigation, such as the extended Kalman filter_EKF, the unscented Kalman filter_UKF, the ensemble Kalman filter_EnKF and the particle filter_PF, and develops the application of ensemble Kalman-particle filter_EnKPF for the AUV integrated navigation. Then through the numerical simulations and experimental data processing, the performances of these integrated navigation algorithms are analyzed and compared.The paper also develops an open source software platform used for AUV control which is named as MOOS_{Mission Orientated Operating Suite}, achieves the functions that the external sensor devices can collect data on MOOS. Finally, the real-time EKF-based navigation on MOOS is developed, and its feasibility in practice also is verified.