Research On Intelligent Navigation Decision Fusion Method Of Upper Stage

With the development of mission requirements and related technologies,the upper stage not only has the function of launch vehicle,but also is capable of autonomous orbital maneuver,long flight time and multiple ignition operations.Its work is no longer limited to the launch and deployment of satellites,but plays an important role in deep-space exploration,on-orbit service,space attack-defense and other space missions.The method of traditional navigation strategy formulation is very effective for the object with a single working mode.However,for the increasingly complex and diverse task frameworks and working modes at the upper stage in the future,traditional methods are difficult to meet the needs of multi-task adaptability.Traditional methods are difficult to meet the needs of multi-task adaptability,do not have the consideration and adaptation to the environment,and do not have the ability to make intelligent decisions.Aiming at the deficiency of navigation system under traditional mode,this paper proposes to design the upper stage intelligent navigation decision algorithm with intelligent method,so as to meet the requirements of the upper stage multi-task applicability and high-precision autonomous navigation in the future.Both neural network and expert system,intelligent algorithm in the realization of decision function requires human experts to provide the corresponding decision basis.In this paper,a navigation strategy database is established by referring to relevant literatures at home and abroad as well as the projects of upper stage successfully completed in recent years,which provides a theoretical basis for decision-making for intelligent methods.Aiming at the problem of slow speed of retrieval and difficult modification of database,this paper uses the Structure of B+ tree to construct the database,which improves the speed of retrieval and operability of database.In view of the traditional design method of navigation scheme can’t meet the requirements of multi-mission adaptability of upper stage,this paper uses the method of combining the theoretical research and numerical simulation,and designs an intelligent navigation decision system based on expert system,neural network and other intelligent methods.This method changes the original design mode,replaces the work of human in the loop,and gives the system the ability to intelligently assign the work of navigation sensors according to the task type and sensor configuration.Through computer numerical simulation,the performance of the algorithm is verified,and a feasible and effective working idea is proposed for the construction of navigation system of the upper stage.The upper stage studied in this paper is the upper stage which can be used for various types of tasks.In the process of carrying out different space missions,the switch of sensor and the change of navigation configuration will inevitably occur.It will produce nonlinear noise which is difficult to be ignored in navigation system and affect the accuracy of navigation system.To solve this problem,this paper uses ANFIS to modify the filtering algorithm.This algorithm solves the random non-Gaussian noise caused by the switch of sensor or the change of navigation configuration in the working process of the upper stage,and ensures the navigation accuracy of the integrated navigation system.