Rapid and Accurate Ins Alignment for Land Applications

Kai-Wei Chiang,Yun-Wen Huang,Xiaoji Niu
DOI: https://doi.org/10.1179/003962610x12747001420348
2010-01-01
Abstract:Abstract INS has been widely used in applications such as aerospace guidance and land vehicle navigation. Position, velocity and attitude are navigation parameters which need to be determined on a moving platform by processing the measurements of inertial sensors including specific forces and angular rates. Generally speaking, the initial attitude angles between the body and navigation frames play an important role, because they could affect the accuracy of the navigation solutions. Therefore, those angles have to be estimated accurately prior to switching the INS into navigation mode using a technique known as INS alignment. In the fine alignment, an optimal estimator such as the Kalman filter is used to perform the alignment process. Due to measurement errors, the Kalman filter takes about 10 to 15 minutes to converge. Those errors increase the alignment time and deteriorate the overall accuracy of the initial attitude angles estimated. In this article, a hybrid alignment scheme which integrates an ANFIS with the Kalman filter is proposed to obtain a better accuracy of the initial attitude angles with the emphasis on the heading angle. This method reduces the time required to achieve reasonable results for fine alignment. A tactical grade inertial measurement unit is used to verify the performance of the proposed scheme in this study. The preliminary results presented in this article indicate that a faster alignment procedure with higher accuracy can be achieved through the use of proposed scheme.
What problem does this paper attempt to address?