Rotary INS self-alignment method based on backtracking filtering under large misalignment angle

In the initial alignment, the accuracy of static base alignment is mainly limited by the short alignment time, the poor observability of azimuth misalignment, and sensor error. Traditional alignment methods cannot take into account the alignment time and accuracy without relying on external measurement assistance. In order to solve these problems, a new alignment method for RINS (Rotary Inertial Navigation System) based on BKF (Backtracking Kalman filtering) is proposed in this article. Firstly, innovative rotation modulation technology is introduced into the initial alignment process to eliminate the influence of sensors bias and improve the observability of azimuth misalignment. Secondly, the first forward filtering constructs KF for optimal estimation while storing the data from RINS into computer. Finally, the result of the last optimal estimation is used as the initial, and proposed scheme performs multiple forward filtering and backtracking filtering processes. Then, an accurate attitude will be obtained quickly. The alignment accuracy of the proposed scheme under large misalignment angle was evaluated through simulation test and shearer alignment experiment. Comparing with other effective static base alignment schemes, the accuracy of azimuth misalignment angle in the proposed scheme was improved at least 65%, and alignment duration of proposed scheme reduced to 85% of these schemes. The experiment has proven that proposed scheme is a high precision self-alignment method suitable for underground shearer, which can obtain accurate initial attitude under short time and large misalignment angle conditions without the assistance from external measurement.