Research On Critical Technology Of High Performance Navigation Method For Aerospace Vehicle Based On Dual Quaternions

Aerospace vehicle has the characteristics of large flying airspace,wide speed range and complex missions.And in different airspace and speed domain,the mission and availability of the navigation system are not always the same.While the existing navigation methods are applied to aerospace vehicle,there are still some shortcomings in the aspects of accuracy,calculation efficiency and reliability.Therefore,it is necessary to design corresponding navigation methods according to the environmental and motion characteristics of the aerospace vehicle during different flight period,so as to meet the strict requirements for autonomous,reliable and accurate navigation system.Beacause reentry period,terminal area energy management period and in-orbit period are typical flying periods of aerospace vehicle.In this paper,aiming at these three periods,a high-dynamic strapdown inertial navigation algorithm,a SINS/GNSS in-flight alignment method under large misalignment angle and a SINS/GNSS/CNS integrated navigation method have been researched respectively.Aimimg at the problem that aerospace vehicle can only rely on inertial navigation in reentry period,but the traditional strapdown inertial navigation algorithm has low precision in high-dynamic environment,this paper has researched on a high-dynamic strapdown inertial navigation algorithm based on dual quaternion.With geographic frame being the navigation frame,calculation formulas of the navigation parameters in geographic frame are derived and the optimal compensation method for strapdown inertial navigation algorithm based on dual quaternion has been researched.Finally,from the aspects of theoretical analysis and simulation verification,the superiority of the strapdown inertial navigation algorithm based on dual quaternion has been compared with that of the traditional algorithm in high-precision and high-dynamic environment.Aimimg at the problem that in the initial stage of terminal area energy management period,the inertial navigation system is at large misalignment,but the alignment efficiency of the existing nonlinear in-flight alignment algorithm is low,this paper has proposed a SINS/GNSS in-flight alignment method based on dual quaternion under large misalignment angle.By deriving the nonlinear error equation,the nonlinear filter model of the SINS/GNSS has been established and a simplified UKF nonlinear filtering algorithm has also been researched.On the basis of analyzing the influence of nonlinear and linear filtering model on the performance of in-flight alignment,UKF and KF are both used to work at different periods for alignment,which effectively improves the accuracy of alignment and reduces calculation.Aimimg at the problem that the SINS/CNS in traditional SINS/GNSS/CNS has low precision,and the chi-square failure detection method can not determine the end time of the fault in the case of long time failure of GNSS during in-orbit period,this paper proposed an improved SINS/GNSS/CNS integrated navigation method based on dual quaternions.Using the observability analysis method to reduce the dimension of the local filters,an improved SINS/GNSS/CNS federated filter based on dual quaternions has been designed,which improves the real-time performance and navigation accuracy at the same time.The SINS and CNS are integrated directly by inertial attitudes,which makes up the defect of the traditional SINS/CNS.Besides,calibrating the inertial device errors on-line and using these to compensate SINS during the GNSS fault period can effectively improve the performance of chi-square fault detection method.Finally,a simulation test platform based on STK/Matlab/RobotFramework to test aerospace vehicle's strapdown inertial navigation algorithm and navigation information fusion method has been bult.The platform is easy to maintain and redevelop.It can simulate the flight trajectory of the high fidelity for aerospace vehicle,automatically determine whether the simulation test is passed according to the set conditions,and generate the simulation test report that is illustrated.It was used to test the algorithms and methods proposed in this paper,and the test results show that the proposed methods was more effective than the traditional ones,which provide new effective reference methods for aerospace vehicle to achieve high navigation performance.