Redundant Configuration And Fault Management Scheme Of Strapdown System

With the development of technology, missile accuracy required in the military are increasing and also missiles are required to own high reliability in a variety of environments, because it is directly related to the success of the mission. The reliability and accuracy of navigation systems depend on failure rate and accuracy of the components within the system. In general, the inertial sensors of the SINS have lower reliability and accuracy than other components so that the increase in the sins of the reliability and accuracy is essential.This paper works on the system, which has a redundancy of four gyros and four accelerometer redundant strap-down systems, for the distribution structure optimization and fault management program design. For the four gyro components in the scheme of configuration structures, three installation methods are proposed: orthogonal and oblique configuration and two cone oblique configurations. Fault management program is a combination of fault detection and fault isolation.When redundancy configuration of SINS is in discussion, we introduce the performance index function. Through optimal analysis of the three configuration based on the improved function, it is known that cone oblique configuration is better than orthogonal oblique configuration in the reliability.In the fault management design, combining the characteristics of four top components,we design a Generalized Likelihood Ratio fault detection of four-point detection program and Linear prediction program. The simulation results show that the combination of two methods can detect and isolate the fault gyros to enhance the reliability and accuracy of the system. During fault detection, different detection threshold will affect the performance of fault management program. Therefore, a reasonable detection threshold is an essential factor in the performance of fault management program.Taking into account of general output of the inertia device containing the installation error, calibration error and constant coefficient of deviation, these errors in the missile maneuvering state would affect the performance of fault detection. To solve the problem, we design a Kalman Filter using the Odd-even Vector compensation package, which can be proved effective through the simulations which compare the situation with or without fault and the fault in different magnitude.