Design Of Micro Attitude Reference System Based On IMU/GPS

With the rapid development of unmanned aerial vehicle(UAV)technology,precise attitude determination,which is one of the key technologies of UAV,is becoming more and more important.At present,inertial navigation system and magnetic sensor are the main methods of attitude determination.The tracking performance of attitude determination system is weakened and even the danger of plane crash under the bad environment such as chaotic flight,so this paper designs a set of practical and stable attitude measurement system under the traction of horizontal project.The main work of this paper is as follows: the motion characteristics of UAV during flight are analyzed,and the output model is established according to the error characteristics of selected inertial sensors,magnetometers and GPS Global Positioning system.For a variety of physical information fusion design to pave the way.Aiming at the motion characteristics and selected sensors of UAV,an attitude determination algorithm based on complementary filter and Kalman filter is designed.This algorithm improves the defects that the traditional complementary filtering is not suitable for high dynamic environment and the traditional Kalman filter can not accurately and quickly measure the heading angle.The hardware platform is the basis of the algorithm.According to the project requirements,the hardware platform with DSP FPGA as the core processor is built to realize the high-speed acquisition and storage of sensor data,as well as the real-time transmission and storage of the solution data.Based on the kinematics of the UAV,the flight simulation experiments are carried out by the trajectory generator and the semi-physical simulation experiments are carried out,respectively,which use the turntable to simulate the large swing experiment in the airborne environment and the vehicle experiment in the outdoor bumpy environment.Finally,the experimental results show that the system can keep high precision output in both environments and meet the requirements of the project.The main contribution of this paper is to propose an attitude determination algorithm based on MEMS sensor,which combines Kalman and complementary filtering.Based on the analysis of the overall structure of the paper,this paper expounds the significance and research background of this subject,describes the important role of attitudeand navigation system in the present scientific and technological era,and briefly introduces the basic principles of inertial navigation.The error model of MEMS sensor is analyzed and the Kalman filter model is established.By analyzing the basic principle and characteristics of complementary filter and Kalman filter,a fusion algorithm based on complementary filter and Kalman filter is proposed.The accuracy of the heading attitude system is evaluated by the high precision three axis turntable and the high precision dynamic positioning and orientation system,and the practicability and reliability of the heading attitude system are verified.