Underwater Vehicle Measured System And Data Processing Method

During the research of underwater vehicle, in order to monitor the state , know about the performance and problem, ameliorate and optimize design project, associating with the actual engineering project, a inside measurement recording system is designed for measuring, memorizing and control, and the beforehand disposal research of testing data is done. The main contribution of this thesis is summarized as follows.An inside measurement and recording system is designed according to the request of engineering project. Its basic composing principle, makeup of hardware and the design of interface communication are introduced in detail.To eliminate the exceptional value of measured data, 3δ rule and other methods are studied. The problem to synchronization of sensors is resolved by using 'interpolation' or 'fit'.Inertial measuring unit (IMU) misalignment and gyros drift are one of the error sources to the parameters of the strap down navigation .In order to solve this problem; the paper forms a new filtering model using error equations of velocity and attitude matching and error equations of misalignment. Estimation of IMU misalignment and gyros drift are done quickly and effectively to use improved BP neural network.Attitude Algorithm Based on Kalman Filtering is brought forward based on the quaternion differential coefficient and the domestic and overseas rotation vector attitude algorithm, utilizing Kalman filtering to estimate optimal of experimental data.After many experiments and data analysis, attitude angle and navigation simulation track curves of underwater vehicle are gained. It validates the function and efficiency of entire system well by contrasting with the reality. The result of experiment also shows that algorithm in this paper can used in realism perfectly.