Research On Optimization Algorithm Of UAV Attitude Based On Bionic Polarized Light

Navigation technology is an important symbol of the development level of national science and technology.It plays an increasingly important role in modern society,from the field of civil use to the field of national defense equipment.Traditional navigation information usually comes from electromagnetic field,geomagnetic field,satellite signal and other controllable factors,and its working environment is limited in many aspects.Using polarized light in the earth's atmosphere for navigation and orientation has the advantages of complete autonomy,strong concealment and high stability,which has been widely concerned by scholars at home and abroad in recent years.Based on the bionic polarized light,this paper realizes the attitude determination of small UAV,designs and improves the algorithm of attitude determination based on the theory of unconstrained optimization,and provides a new idea for bionic polarized light navigation.In this paper,firstly,the frame of navigation coordinate system is established,the different representation methods of vehicle attitude are analyzed and compared,and the mathematical model of attitude quaternion matrix is constructed.The theory of Rayleigh scattering and the mode of atmospheric polarization distribution are introduced.Based on this,the principle of polarization information acquisition is analyzed in detail,and the theoretical measurement model of polarized light is constructed.Using the open source community of paparazzi UAV,a small fixed wing UAV navigation experiment platform is designed and built,and its overall architecture,software and hardware configuration scheme are analyzed.The polarized light sensor and the optical flow sensor are extended by using the external interface reserved by the flight control system.By defining the configuration file and improving the flight control code,the software driving control of the two bionic navigation sensors and the data display of the ground station are completed.The accuracy of the important sensors used in UAV navigation is calibrated,and the UAV autonomous flight experiment is completed by optimizing the flight control parameters.The results show that the original attitude solution algorithm of the flight control system can be used as a reference for subsequent research.In view of the shortcomings of the existing polarizing light assisted navigation attitude determination methods,based on the gradient optimization in the unconstrained optimization method,this paper introduces real-time Quaternion Differential to build a dynamic step factor to predict the attitude update trend;designs conjugate parameters to ensure that the front and back search directions are conjugate with each other during the attitude iterative optimization process,so as to improve the convergence speed;builds a fusion gyroscope and acceleration The conjugate gradient optimization model is used to calculate the attitude of the sensor.The experimental results show that the method proposed in this paper has good static stability and accuracy,and can accurately track the attitude change in dynamic environment.The comparison with the gradient descent method shows that the convergence speed of the method is faster,which effectively guarantees the accuracy and reliability of the polarized light assisted attitude reference system.In order to realize the deep fusion of sensor data,an EKF cascaded filtering method based on bionic polarized light attitude calculation is proposed.The experimental results show that the cascaded filtering method not only has good static accuracy and stability,but also further improves the attitude calculation performance in dynamic environment,and reduces the overshoot lag phenomenon when the attitude angle changes abruptly.On this basis,a variable noise adaptive filtering model is constructed to suppress the interference of non gravity acceleration,and the dynamic estimation of non gravity acceleration noise is realized.The experimental results show that this method improves the adaptive ability and robustness of the polarized light attitude reference system in strong maneuvering environment,and can provide more reliable parameter information for subsequent navigation solutions.