The Design And Implementation Of Data Buoy Posture Sensor Based On MEMS

Data buoy plays a crucial role in marine hydro-meteorological forecast, marinerecourse development and marine disaster warning. It can continuously monitor themarine environment and climate for a long time. Data buoy should measure itsattitude information to calibrate and compensate measured data of wind speed, winddirection and ocean current.With the comprehensive development of Micro-electromechanical Systems,manufacture of the low-cost, low power and small posture sensor based on MEMSgradually becomes the development trend in the field of attitude measurement. Due tothe attitude divergence and attitude error from MEMS sensor, we need to utilizemulti-sensor to obtain redundant attitude information, and use data fusion technologyto improve the measurement precision and stability of the system.According to the above requirements and problems, the paper designs andmanufactures a kind of data buoy posture sensor based on MEMS. The system adoptsSTM32F103microprocessor which owns rich on-chip resources as the hardwareplatform. The system utilizes three-axis accelerometer, three-axis magneto-resistivesensor and three-axis gyroscope to obtain the redundant attitude information. Theextended kalman filter is used for information fusion to prevent attitude divergenceand attitude error. The system transmits the final attitude information to PC via aserial port. The system adopts μC/OS-II real-time operating system which ownspublic source code as the software platform and successfully transplants μC/OS-IIreal-time operating system into STM32F103microprocessor. On the basis of softwareplatform, the system designs hardware drivers and user applications. After the systemstable operation, we analyze the various causes of the attitude error in the system andtheir compensation method, and verify the feasibility of ellipse hypothesis for thecompensation of magnetic heading.According to the system test, the system based on STM32F103microprocessorand μC/OS-II real-time operating system is composed of three-axis accelerometer,three-axis magneto-resistive sensor and three-axis gyroscope. After FIR digital filter,extended kalman filter and error compensation, the measurement precision of thesystem can meet the requirements of data buoy. And the system compared withtraditional posture sensor used in data buoy is small, low-cost and high real time. The system is of great research value and application value.