Terrain Features Based Autonomous Navigation And Guidance For Planetary Landing

With the increasingly complex ways of deep space explorations,landing explorations is becoming a more advanced stage of development of deep space exploration mission,in order to obtain higher valued scientific data,future planetary exploration mission require the detector has the precise landing ability in complex terrain environment,which needs navigation and guidance system higher requirements.The abundant terrain features of the planetary surface provide a rich source of information for the land navigation detection and guidance system,which makes the autonomous optical navigation and guidance based on the catalog feature become important technologies for the development of the future planetary landing mission.In this paper,the method of autonomous navigation and guidance for planetary landing based on star-table features is studied for safe landing of complicated terrain on planetary surface.The main work of the paper is as follows:Firstly,for the problem of acquisition and processing of planetary landing optical navigation information,the feature extraction method of star catalog based on image information is studied.Firstly,the features of the craters in a single image are preprocessed by using the feature detection method based on SVM.On this basis,the feature points of the image are classified and analyzed by cluster analysis method.And the problem of multi-crater mislabelling in the traditional crater detection method is solved.Secondly,the autonomous optical navigation scheme of the planetary landing is designed based on the star feature information.Firstly,the planetary landing coordinate system is established,and the navigation measurement model is built on this basis.On this basis,the observability of the navigation system is analyzed,and the optimization scheme of navigation roadmap based on star table feature is given.The landing navigation filter is designed based on extended Kalman filter.Thirdly,aiming at the problem of safe landing of complex terrain on the surface of the planet,the obstacle detection and safe landing guidance strategy are studied.Firstly,this paper studies the detection methods of different obstacle types,and gives the safety evaluation criteria of planetary terrain based on the statistical information of feature distribution,which can make up for the defects of traditional obstacle detection methods in multi-type complex obstacle terrain detection and evaluation.On this basis,The strategy of safe landing guidance is designed based on Lyapunov function.Finally,based on the method of stochastic terrain generation,the terrain feature model base of the planetary surface was established.On this basis,the synthetic simulation environment of the planetary landing and guidance was constructed.The accuracy of the navigation roadmap,initial state deviation,light condition,terrain distribution The performance of the landing guidance and guidance method is simulated and analyzed,and the problems that need attention in the design of the planetary landing navigation scheme and the guidance strategy based on the star table feature are given.