Systems Engineering and Electronics ›› 2021, Vol. 43 ›› Issue (5): 1310-1315.doi: 10.12305/j.issn.1001-506X.2021.05.19

• Guidance, Navigation and Control • Previous Articles     Next Articles

Initial alignment method based on information reuse and algorithm fusion

Zhihao XU1(), Zhaofa ZHOU1,*(), Zhenjun CHANG1(), Qi GUO2()   

  1. 1. Missile Engineering Institute, Rocket Force University of Engineering, Xi'an 710025, China
    2. Unit 96902 of the PLA, Beijing 100015, China
  • Received:2020-07-01 Online:2021-05-01 Published:2021-04-27
  • Contact: Zhaofa ZHOU E-mail:1500105885@qq.com;zhaofa_zhou@163.com;changzj2105@163.com;Guoqi_academic@163.com

Abstract:

Through theoretical analysis of the error characteristics of the multi-vector attitude determination method and the observability of the optimal estimation initial alignment, it is concluded that the accuracy of the two methods under a static base is equivalent. This conclusion is verified through experiments, and it is found that when the alignment time is short, the multi-vector attitude determination method has large fluctuations. Therefore, in order to improve the stability of the algorithm, all data in the alignment phase is used for multi-vector attitude determination. And the optimal estimation alignment is extended to the entire alignment stage to improve the alignment accuracy. Combining the characteristics of the two algorithms with the same limit accuracy but different anti-jamming capabilities, a vehicle-mounted alignment scheme with algorithm fusion is designed. The vehicle experiments show that the new scheme improves the accuracy and robustness of the initial alignment under interference conditions, and the standard deviation of the azimuth angle is reduced from 1.48′ in the traditional scheme to 0.72′.

Key words: strapdown inertial navigation system (SINS), initial alignment, information reuse, algorithm fusion

CLC Number: 

[an error occurred while processing this directive]