1. A High-Order Enhanced Attitude Algorithm Under Angular-Rate Input Condition
- Author
-
Ding Peng, Cheng Xianghong, and Wang Yineng
- Subjects
Computer science ,Computation ,010401 analytical chemistry ,ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION ,Gyroscope ,Rotation matrix ,01 natural sciences ,0104 chemical sciences ,law.invention ,Vibration ,symbols.namesake ,Inertial measurement unit ,law ,Taylor series ,symbols ,Electrical and Electronic Engineering ,Quaternion ,Instrumentation ,Algorithm ,Inertial navigation system - Abstract
With the development of high-precision strapdown inertial navigation systems (SINS), it is increasingly important to improve the accuracy of navigation algorithms to match high-precision inertial sensors and meet the navigation localization requirements in high dynamic environments. However, it is difficult to further improve the accuracy of traditional second-order angular rate-based attitude algorithms due to the neglect of the triple-cross-product term and the limitation of attitude updating frequency. In this paper, a high-order rotation vector-based coning algorithm with compressed form for high accuracy attitude computation of online processing systems is proposed. The theoretical third-order terms of rotation vector differential equation in pure coning motion are deduced in detail and the angular rate vectors of previous attitude updating cycle are additionally used to estimate angular increment, coning correction term and the third-order terms. The error analysis and optimization of the coefficients for proposed algorithm are conducted by utilizing the Taylor series expansions in powers of coning frequency. Simulations under coning vibration environment and turntable experiments in laboratory were performed to verify the performance of the algorithm. The results show that the proposed algorithm can obviously reduce the noncommutativity error and obtain higher accuracy compared with the existing angular rate-based algorithms.
- Published
- 2021