Research Of UAV Integrated Navigation Algorithm Based On Atmospheric Polarization

Navigation technology is one of the key technology in unmanned aerial vehicle (UAV) system, with the development of science and technology and the diversity of applications, the existing navigation technology appear a great deal of defects, therefore, more and more scholars shift the research focus to the bionic navigation technology, the existing research results show that the bionic navigation technology has strong anti-interference ability, high concealment and the advantages of security and stability. The research of this thesis is three-dimensional navigation with bionic polarized light, focuses on multi-sensor information fusion algorithm and construction of UAV experimental facilities, aimed at improving the stability and practicability of the polarized light integrated navigation algorithm.This thesis introduces the basic concept of the unmanned aerial vehicle (UAV) navigation and inertial sensor theory model and error model firstly, explains attitude cosine matrix method through the detailed derivation process, then uses Matlab to construct platform for unmanned aerial vehicle (UAV) navigation, this platform includes pure simulation environment and the measured data simulation environment, the simulation results are consistent with the intended effect has proved the validity of the platform, provides the support for further research topics.Due to polarized light navigation has 180° fuzziness of measurement and the standard error model of polarized sensor has not being established, this thesis put forwards the MIMU/POL complementary filter integrated navigation algorithm, the proposed algorithm meets the requirements of the three-dimensional navigation and inhibit the divergence angle of Unmanned Aerial Vehicle effectively and less calculation which has better practicability. On this basis, this thesis studies the impact of abnormal date in polarization sensor measurements to attitude estimation caused by environmental interference and proposes a method of adjusting complementary parameter dynamically, the simulation results show that the proposed method can eliminate the influences of outliers in the measurements of polarization sensor effectively, which will bring about the error of attitude estimation.In addition, this thesis develops the UAV experiment platform based on Paparazzi UAV open source project and completes autonomous flight through modifying the configuration file and adjusting the control parameters. Apart from that, this thesis analyses the software framework and the methods of hardware management. On this basis, the thesis explains the access technology of polarized light sensor in details. The MIMU/polarized light complementary filter algorithm is assessed by using the data measured by sensors, experimental results demonstrate that the proposed algorithm is able to solve the problems existing in the polarized light navigation and export the accurate attitude with long-term stability to UAVs.