Xu Bo

Analyse of damping network effect on SINS

Inertial Navigation System (INS) is actually an unstable system. Affected by some kinds of error sources, such as inertial measurements sensors, initial misalignments, INS outputs contain three kinds of periodic oscillation errors. It is a very effective way to add damping network in the Platform INS. This method can also be used in the Strapdown INS (SINS). The advantage of using this method is that the periodic oscillations errors can be damped independently without external information. This paper analysis the damping network effects on the outputs of SINS under static condition. The simulation results show that using damping network could eliminate the periodic oscillation errors effectively, but have no effect on the constant and drift errors for SINS.

The observation analysis of the initial alignment in the warship inertial navigation system

The initial alignment is one of the key technologies in the inertial navigation system. Nowadays Kalman filter (KF) is always used in the initial alignment to evaluate the angle error, the drift of the gyroscope and the bias of the accelerator. Before the design of the KF, we have to do the observation analysis of the KF to determine the effect of the filter. At first, this paper establishes the error model based on the moving base, then uses the singular and the spectrum condition number to make the comprehensive observability analysis of the subsection continuous system both on the static base and the moving base. Then through the reduction of the error equations, we analyze the observability in the theory. We did the simulation and also make the comparison.

A Fast and Optimal Method Based on Fuzzy Neural Networks for SINS Initial Alignment

The paper presents an approach which applies LQ optimal control in Strap-down Inertial Navigation System (SINS) initial alignment. The detailed analysis process about this method has been shown, and anti-jamming designs were done for the approximate step-wise disturbance which appeared in the system, to gain the optimal control parameter, the fuzzy neural network is used in the method. Simulations are performed to validate the scheme for initial alignment, obtained the same precision to other alignment method. And the most splendid was that the using of the LQ optimal control reduces the alignment time consumedly much more quickly than other traditional method. What is discovered in the simulation is while the initial error angles were smaller than 10 degrees; error angle will converge to quite good precision in several seconds.