An error-free method for updating values ​​of strapdown inertial navigation

Abstract

The invention discloses an error-free strapdown inertial navigation value updating method, which relates to the technical field of strapdown inertial navigation. According to the attitude array, velocity and position differential equation of the strapdown inertial navigation, the Taylor series expansion method is directly adopted to give A complete set of strapdown inertial navigation update numerical method is proposed. Under the condition that the output angular velocity of the gyro and the output ratio of the accelerometer satisfy the assumption of polynomial form, this method does not have any principle error, which implies non-exchangeable errors for attitude, velocity and position updates. precise compensation. For the convenience of theoretical research and comparative analysis, the present invention uses the inertial coordinate system as the navigation reference system, and provides a concise method for updating the value of the strapdown inertial navigation. Finally, it is verified by simulation experiments that there is no principle error in the method, and it has high calculation accuracy.

Description

technical field [0001] The invention relates to the technical field of strapdown inertial navigation, in particular to an error-free strapdown inertial navigation value updating method. Background technique [0002] Since the 1960s and 1970s, the traditional algorithm of SINS has been developed for decades, and its mainstream design methods and details have been completed. Attitude update is the core of the whole strapdown algorithm. In order to compensate the attitude non-exchangeable error, based on the concept of equivalent rotation vector, multi-sample and its optimization algorithm are proposed. Using the principle of duality, the attitude multi-subsampling algorithm can be directly applied to the velocity update algorithm. Compared with the former two, the influence of the algorithm error of the location update algorithm is relatively small, and generally can not be considered in many applications. [0003] However, the traditional multi-sample algorithm for non-exch...

AI Technical Summary

Problems solved by technology

[0003] However, the traditional multi-sample algorithm for non-exchangeable error compensation is derived on the basis of the second-order approximation of the equivalent rotation vector equation (Bortz equation), which essentially has a principle error, especially in large maneuvering environments. Neglect, sometimes the error of the high-sample algorithm is larger than that of the low-sample algorithm

Even if the consideration is more thorough and the high-order approximation of the Bortz equation is retained, there are still principle errors, and the derivation process of the high-order compensation algorithm will become very complicated

Images