| The rotor UAV has the characteristics of compact structure,high flexibility and flexible operation.It can be equipped with a variety of devices to achieve the function of vertical takeoff and landing,and hanging over the air,and it's Widely used in military,agriculture,film and television,and other fields.But in order to achieve the smooth flight of UAVs,complex sensors,fast controllers and actuators,and complex navigation control algorithms are required to support the system.However,the position and attitude information,like the UAV's eyes,is the basis for its autonomous flight.As a multi-sensor data fusion system,INS/GPS integrated navigation system can get high precision navigation information.In this paper,based on two important topics of the project team,Extensive research has been carried out for correlation algorithm of attitude measurement and integrated navigation for UAV based on multi-sensor data fusion.The main work is listed as follows:(1)Aiming at the system's mission requirements and work environment,the hardware and software are designed,and the rotor UAV experimental platform is established.The performance of relevant hardware modules is analyzed;The flight control board is designed using the Altium Designer software;The upper computer is developed by the Labview software which is used to receive,display and analyze the experimental data.And the flight control program is wrote using the Keil sof-tware.The system control structure block diagram and software work flow chart is also designed.(2)As the theoretical foundation of the full text,firstly,the complementary filter algorithm is studied deeply.The basic principle of the complementary filter is analyzed from the view of time domain and frequency domain.An improved method of complementary filter is proposed,by adjusting the relationship between the complementary filter coefficient and the filter frequency,when the noise characteristics of the system are changed.Then we study the standard Kalman filter and extended Kalman filter applied to different objects.By analyzing the correlation coefficient and the deep meaning of the matrix,the five core formulas of Kalman filter are deeply analyzed.The improved method of model error and noise uncertainty is summarized to improve the robustness and accuracy of the filter.On this basis,the multi stage loosely coupled data fusion scheme are determined,which based on nine navigation parameters such as attitude,position and speed.(3)In this paper,the basic principle of quaternion attitude calculation is studied.Then the quaternion attitude calculation method based on complementary filter.is designed,which use the quaternion to update the attitude.By using the relationship between acceleration and attitude angle,an extended Kalman filter is used to design a new acceleration attitude solve.Taking into account the influence of the non-gravity acceleration of the UAV,the adaptive improvement of the noise is made to the filter.Finally,the feasibility of the two algorithms is proved by the comparison experiment,and the latter is better than the former.(4)A simple and practical integrated navigation algorithm based on linear Kalman filter is designed.The theory of the algorithm is simple,and it has less computation and is high in real time.Considering the impact caused by the time-varying characteristics of the system noise in terms of statistical characteristics.The method of linear iterative to adjust the observation noise parameters is designed in order to improve the filter adaptively.At last,the algorithm is verified on the experimental platform of the large rotor UAV and the experimental platform of the micro rotor UAV.The results show that the algorithm can meet the demand of the UAV to hover in the precision range of lm outdoors and 20mm indoors. |
PDF Full Text Request