Study On Attitude Measurement Based On DSP

Today,with the rapid development of technology, the people are paying more and more attention to navigation technology in the world. Especially in the aviation, aerospace, satellite, missile and other fields. Among them,getting the exact attitude parameters information for the missiles is an important task in the navigation. In this paper, to further improve the accuracy of calculating attitude parameters,the magnetometer and MEMS Gyro attitude testing system is put to use.The improved information fusion Kalman filter algorithm is designed, and suppresses the system error effectively. Continue to study the subject, this paper provides a certain reference value. The specific work in the following aspects:Firstly, the basic principles of inertial navigation systems and attitude parameters of solution method are introduced in detail, and describes the characteristics of the geomagnetic field and the basic theory briefly.Secondly, in order to enhance the calculating attitude parameters capacity for attitude measurement system, DSP (digital signal processor) as the key components, to achieve the IMU (inertial measurement unit) signal acquisition, storage, calculating attitude information of the hardware system is designed.Finally,the paper analyses quaternion solving method of gyroscope attitude determination and the nonlinear least square method of magnetometer attitude determination. On the basis of System error statistics, the attitude parameters resolving method is designed by means of using BP neural network to correct the coefficient of Kalman gain matrix, and the algorithm is validated by non-magnetic three-axis turntable. At the same time, compared with the conventional kalman filter algorithm, the Calculation results demonstrate that the improved Kalman filter algorithm can further inhibit the accumulation of attitude measurement system error, so as to achieve the objective of improving the precision solver. The Solution of attitude parameters is more appropriate information fusion algorithm.