Gaussian particle filtering data processing method based on error quaternion three-dimensional vector distribution

Abstract

The invention provides a Gaussian particle filtering data processing method based on error quaternion three-dimensional vector distribution, which belongs to the technical field of digital filtering and multi-sensor data fusion, and is mainly used for solving the problem of huge calculation burden of a particle filter in attitude estimation. The method takesGaussian particle filtering as a frame,data of a gyroscope, an accelerometer and a magnetometer are fused, and tGaussian distribution of an error quaternion three-dimensional vector on a three-dimensional Euclidean space is taken as posteriori distribution of an attitude estimation error. According to the indirect filtering method taking quaternion as a global attitude parameter and the multiplicative error quaternion as a local attitude error description, normalization of the quaternion in the filter can be guaranteed, sequential estimation of the filter can be achieved, parallel computing can be achieved, the computing speed is increased, the filtering precision is stable, and the method is suitable for attitude estimation, data fusion and other application occasions.

Description

technical field [0001] In order to solve the problem that the particle filter algorithm has a huge calculation burden in aircraft attitude estimation, the present invention proposes an aircraft indirect attitude estimation algorithm based on Gaussian particle filter, which belongs to the technical field of data processing and multi-sensor data fusion, and is suitable for attitude Estimation, data fusion and other applications. Background technique [0002] For attitude estimation, the use of sequential vector measurements has been intensively studied in MAV navigation systems. At present, the mature scheme of attitude estimation is based on the vector observation of gyroscope, accelerometer and magnetometer to determine the attitude of the aircraft. Quaternions, as a globally non-singular minimum-dimensional pose description form, have been widely used in pose estimation problems. In recent decades, a lot of research has been carried out using various new algorithms to imp...

AI Technical Summary

Problems solved by technology

Although QPF is closer to the real value for nonlinear and non-Gaussian problems than other algorithms, in order to overcome the defects of particle degradation and sample scarcity, QPF needs to resample the particles, which increases the computational burden as the number of particles increases.

In recent decades, various scholars have made many improvements to the defects of particle filtering. Although the effect has become better, the calculation has become more complicated.

Images