Autonomous Navigation And Filtering Algorithm Of Small Body Spacecraft In Landing Phase

The "Tianwen-1" Mars spacecraft has been successfully launched,and my China’s small body exploration mission is expected to be implemented around 2024.Compared with large planets and lunar bodies,small bodies are characterized by small size,irregular shape and little a priori information,which require more autonomous navigation capability for their spacecrafts.In this paper,we design a navigation scheme for small-body spacecrafts in the context of small-body landing phases and propose corresponding filtering algorithms to improve the state estimation accuracy of the spacecrafts.The main research of this paper is as follows.First,the coordinate system that the small body spacecraft refers to during the landing process are constructed,and the conversion relations between the coordinate systems are derived.The ellipsoidal harmonic function method is used to model the gravitational field of the landing phase of the small body spacecraft,and then the dynamics equations of the landing phase of the small body spacecraft are established.Secondly,in view of the lack of matching feature points,craters and other landmarks on the surface of small bodies,irregular curves are used as navigation landmarks.basis on successful matching,two-dimensional and three-dimensional information of points and tangents on the irregular curves are used to solve the rotation matrix and translation matrix of the navigation camera coordinate system relative to the landing point coordinate system of the small celestial body,that is,solve the position and attitude of the small celestial body spacecraft.In view of the non-Gaussian white noise and other unknown sudden noise interference that the small body spacecraft may be subjected to during the landing phase,the UFIR filtering algorithm is used to estimate the position and attitude of the small celestial body spacecraft during the landing phase.The simulation is carried out for different noise interference,and the simulation results verify the feasibility of the UFIR filtering navigation algorithm based on irregular curve matching,and its stability is also better than that of the traditional Kalman filtering algorithm.Finally,to solve the problem of limited extraction of navigation landmarks when the surface of small bodies is dark,a combined navigation scheme of Lidar and vision is adopted.Therefore,the rotation of the small celestial body and the position vector of the navigation camera to the feature point on the surface of the small body are added to the filter as a state quantity.In the algorithm,in order to ensure the linearization accuracy of the equation,the UKF algorithm is used to estimate the pose of the small body spacecraft.The Lidar / visual integrated navigation algorithm is simulated with the Eros433 asteroid as the background.the position estimation error is within 0.6 m and the attitude estimation error is within 1.5°under the condition that five feature points are successfully matched.