Intelligent Astronomy Navigation Method For Spacecraft Based On Celestial Body

Autonomous navigation technology plays an indispensable role in the application of spacecraft in deep space exploration and manned landing.It is crucial to reduce the burden on ground-based monitoring and control stations,improve the autonomous survival capability of spacecraft,and expand the mission potential of spacecraft.The current autonomous navigation method mainly uses the method of extracting natural celestial signals for navigation.However,the space environment is complex and diverse,and the available celestial navigation information has a wide range of variations,and it cannot guarantee stable acquisition.Therefore,there is an urgent need to improve the autonomous acquisition and processing capabilities of spacecraft navigation information in complex environments.For these problems,the main research object of this article is the autonomous navigation method with astronomical observation as the navigation information source.In order to improve the spacecraft's ability to extract and process navigation information,and to achieve a higher degree of navigation autonomy,the use of OR-Analysis Methodologies such as Law,Ontology,Semantic Web Technology,Bayesian Networks,Factorial Graphs,etc.,have designed calculation methods for knowledge representation,knowledge reasoning,uncertainty reasoning,and factor graph fusion in the navigation process.The main contents include the following parts:In view of the randomness,heterogeneity,and complexity of celestial navigation information sources,this paper proposes the use of knowledge ontology technology to make inference decisions on information processing processes.Firstly,it analyzes the decision process of navigation information extraction and calculation,and uses OR and graph to reduce,and proposes the inference steps of information nodes and method nodes.Then,ontology technology is used to construct knowledge ontology in the field of astronomy navigation,and OWL ontology description language is used to establish the ontology library of celestial navigation knowledge.The traditional navigation method can only give position and error,and cannot provide uncertainty information such as the uncertainty of the result.In order to calculate the credibility of the navigation results,the established ontology semantic network is transformed into a Bayesian probability network.Using PR-OWL language,a multiobjective Bayesian network for probabilistic analysis was constructed.From the initial confidence probability of each navigation information source,the Bayesian recursion can be used to obtain the credibility of the final navigation result.In order to effectively use the established Bayesian network to fuse stochastic and heterogeneous multi-source navigation information,factor map techniques are used to perform global optimization of the final navigation state.Analyze each information source first,establish the navigator factor node of this information.The factor graph is flexible in construction,the navigation signal can be plug-and-play,and the navigation information source can be increased or decreased at any time.On the basis of the Bayesian network,the key nodes are extracted and converted into a factor graph.Using the incremental smoothing algorithm,global optimization of all state quantities can be solved.Finally,in order to verify the knowledge-based astronomy navigation method proposed in this paper,a navigation method based on the knowledge of ground/month celestial bodies was designed.First simulate the space/month object information that the spacecraft can receive in the lunar orbit.Use this article's navigation method to process and fuse this information.Experimental results show that the navigation method proposed in this paper has the ability to calculate the credibility of navigation information.Under the same information source,the navigation results are better than the traditional Kalman filter fusion algorithm.