Research On The Key Technology Of Multi-source Redundant Fault-tolerant Navigation For Aerospace Vehicle

Aerospace vehicle refers to a new type of aircraft with two functions: spacecraft and aircraft.It is launched from the ground to the orbital space through the booster,which completes the mission and runs in orbit for a long time.It finally arrives at the reentry point and returns to the earth.In addition to emergency or sudden,all operations depend on autonomous navigation control system to carry on the corresponding operation without remote control on the ground.So it will need to design for a autonomous navigation system of complete trajectory.At the same time,autonomy,independence,accuracy and adaptive of the navigation system are required to more strict requirements.At present,most of the navigation methods are designed and implemented according to fixed scenes,which cannot be directly applied to the whole autonomous navigation of aerospace vehicle.Therefore,it is urgent to carry out the research on the realization method of the all-flight navigation system of the aerospace vehicle,design the overall navigation scheme and corresponding navigation algorithm that adapt to the all-flight of the aerospace vehicle,so as to meet the navigation requirements of the allflight of the aerospace vehicle with high adaptability,high precision and high autonomy.Therefore,based on the different flight modes of aerospace vehicle and combined with the requirements of navigation performance in different flight stages,this paper proposed the overall scheme of navigation system,and conducted targeted research on key technologies such as multi-source redundancy fault-tolerant algorithm and others.Paper analyzes the whole flight characteristics of aerospace vehicle,in view of the navigation requirements at different stages of the whole flight,designs the framework based on multi-source information fusion navigation system.On this basis,adapting to each stage of model of navigation system is deduced and the navigation system state equation and measurement equation is established in different stages.Paper designes a navigation system of switching algorithm under different flight phase.And for subsequent multi-source redundant fault-tolerant navigation research provides the basis.A redundant navigation system based on multi-source information fusion is designed to meet the requirement of high reliability of navigation system for long flight of aerospace vehicle.A multi-source redundant fault-tolerant navigation system based on multi-stage filter framework is proposed to simplify the implementation of redundant fault-tolerant navigation system.Aiming at the problem of low sensitivity of traditional redundant navigation system to soft fault,a fuzzy fault-tolerant algorithm based on residual and its rate of change is designed,which has good sensitivity to soft fault and further improves the accuracy of navigation system.In the high-speed re-entry phase of aerospace vehicle,due to the influence of the black obstacle zone,the inertial navigation system cannot effectively obtain the information of other auxiliary navigation systems for error correction,resulting in the continuous accumulation of errors in the inertial navigation system.So it will need to design the method of intelligent navigation based on extreme learning machine.before entering blackout area celestial navigation system can be used to estimate and correct posture error of inertial navigation system.The use of extreme learning machine of learning ability and the ability to predict fast,gets real-time error of inertial navigation system quickly and accurately.Before entering blackout area it must ensure precise attitude information,and effectively improve the performance of the blackout area navigation system.Finally,a simulation verification platform for the all-flight navigation algorithm of aerospace vehicle based on STK is designed and implemented.The platform by STK simulation of high realistic space flight track,can use the platform to simulate test ofnavigation switch algorithm,multiple source redundancy fault-tolerant and blackout area reconstruction algorithm based on extreme learning machine intelligent navigation algorithm.The test results show that the effectiveness of the proposed relative navigation algorithm.And it provides the beneficial reference for high performance navigation of aerospace vehicle.