Research On Autonomous Navigation In The Landing Section Of Small Body Detection Descent

The detection of small celestial bodies is of great significance in revealing the origin of human life and promoting social development.Based on the research background of the descending landing section of the small body exploration mission,this paper combined with the subproject of the national "863" key research and development program "Flexible attached intelligent sensing and Navigation for noncooperative targets" and the subproject of National Defense Basic Scientific Research Project "Research on Optical Feature Extraction of Weak Targets and Development of Optical Navigation System".In view of the problems of insufficient navigation information and low precision during the landing of small celestial body probe,the navigation model establishment,optical autonomous navigation technology,navigation image processing technology and navigation filtering algorithm were studied in depth,so as to provide some technology support and theoretical guarantee for the descent and landing section of the small celestial body exploration mission.The main work of this paper is as follows:(1)The autonomous navigation model of the landing section of small celestial body detection is established.The heliocentric ecliptic coordinate system is used as the reference,on the basis of the pre-established navigation coordinate system,the transformation relationship between each coordinate system is determined,which is convenient to complete the transformation between each coordinate system quickly during calculation.The characteristics of the gravitational field of small celestial bodies and the influence of spatial perturbation are analyzed comprehensively,and the spherical harmonic coefficient method is selected to construct the gravitational field model of small celestial bodies.By analyzing the position and attitude of the probe,the probe dynamics model and navigation camera observation model are established.(2)The autonomous navigation technology based on optical sensor is studied.The optical navigation camera was used to obtain the image data of small celestial bodies in the descending landing stage,and the image coordinates and navigation landmarks of target celestial bodies were determined.The position and pose of the probe were analyzed based on the conditions of three navigation landmarks.(3)Improved navigation image processing algorithm and feature extraction and matching algorithm.Due to the weak environment,uneven illumination and irregular shape of the target object,the surface crater edge information of the celestial body appears insufficient feature extraction and mismatching.Bilateral filtering method and improved histogram equalization algorithm are used for filtering denoising and image enhancement to improve the brightness and clarity of the image while preserving the boundary information of the object.The ORB detection algorithm based on scale invariance was used to extract and match the feature points of the enhanced image to improve the matching rate.(4)Improve the nonlinear navigation filtering algorithm based on classical particle filter.Aiming at the problem of system noise and measurement noise during the trajectory tracking of the detector in descending landing section,an improved particle filter algorithm based on second order central difference method and principal component analysis was proposed to improve the autonomous navigation accuracy of detectors in complex deep space environment.The second order central difference filtering method was used to obtain the optimal importance density function,and the principal component analysis method was introduced to preprocess the collected sample set to prevent particle degradation,reduce tracking error and improve navigation accuracy.