| Deep-sea Work-Class ROV (Remotely Operated Vehicle) is one of the important equipments of human exploration and development of deep-sea environment. The condition of deep sea environment is diverse and complex, High reliability and satisfying work precision navigation system is needed to provide navigation information for Work-Class ROV.Traditionally single type of navigation system can not satisfy the navigation demands of the"Work-Class ROV,so using a various types of information provided by the navigation system,then highly reliable navigation information is achieved through data fusion. It is the mainstream research method in the development of ROV navigation system.This paper constructs the implementation platform of ROV,s navigation system.According to the navigation demands of Work-Class ROV, the sensor is selected according to the precision of navigation system. In the meanwhile, the navigation computer is chosen due to above requirements. In the end, navigation system platform of power supply module, data interfaces module and communication module is completed the design.Secondly, the basic model of the navigation system is established. The error models of inertial devices are built, and the basic principle of SINS system is introduced. The transformation relationship between the coordinate systems, the attitude updating method and the updating equations of SINS system are all derived.The corresponding error models for Electronic compass, Doppler Log and GPS are established.Thirdly, the error sources of MEMS inertial devices are analyzed to establish the error model of inertial MEMS devices, then the Laboratory calibration scheme is proposed and calibration is carried out. Ellipse fitting calibration scheme for electronic compass is derived.Afterwards, the field calibration scheme of navigation system’s sensors is designed and relative experiments are done as well. The initial alignment for the navigation system with MEMS inertial device is studied, and coarse alignment scheme is verified in the experiment.Then, data fusion scheme of ROV navigation system is researched, extended Kalman filter is adopted as the filter of navigation system. Because ROV’snavigation system requires high reliability,several aspects ,which include the filter’s update mode, the navigation system model, sensor’s fault detection and redundant configuration are designed and their simulation experiments are carried out to verify the reliability of ROV’s navigation system.Finally, the integrated navigation system of ROV is experimentally tested. Under outdoor conditions, the ROV’s working environment is simulated, then the experiments of both static and dynamic environment are done separately. Experimental results show that the performance of the navigation system can meet the requirements of the Work-Class ROV. |