Information fusion navigation method for detecting fixed-point soft landing process in deep space

Abstract

The invention relates to an information fusion navigation method for detecting a fixed-point soft landing process in a deep space. An autonomous navigation system of a lander is provided with an inertia measurement unit, a distance measurement sensor, a velocity measurement sensor, an optical imaging sensor and other navigation sensors, and on the basis, in the whole soft landing process, the angular velocity and the non-gravitation acceleration of the lander are measured by the inertia measurement unit, and integration and extrapolation are carried out to provide position and velocity information for the navigation system; slope distance information output by the distance measurement sensor, relative velocity information output by the velocity measurement sensor and feature point pixel coordinate information output by the optical imaging sensor are all processed by Kalman subfilters independent of one another, and the status and error covariance is locally estimated; the local estimation result is subjected to information fusion by a covariance crossing method to obtain the unified position and velocity estimated value of the lander at present, and the unified position and velocity estimated value is fed back to the subfilters for next navigation updating. The method is high in position and velocity estimation precision and good in real-time performance, has fault-tolerant capability and is easy to realize.

Description

technical field [0001] The invention belongs to the field of navigation and relates to an autonomous navigation method for spacecraft based on information fusion. Background technique [0002] The fixed-point soft landing process of deep space exploration takes a short time, the dynamic changes in attitude and orbit are large, the environment is highly unknown, and it is difficult to stop once it starts soft landing on a large celestial body. high demands. [0003] The traditional navigation method based on the ground measurement and control network has the disadvantages of large delay in deep space communication, low accuracy in determining relative position and velocity, and inability to provide detailed geographic information of the landing site. Therefore, the lander must rely on the autonomous navigation system to obtain measurement information through various sensors, and then processed by the navigation algorithm on the lander to estimate its attitude, position, and ...

AI Technical Summary

Problems solved by technology

However, the disadvantage of this method is that part of the navigation accuracy is sacrificed, and the estimation given is not the optimal value.

However, federated filtering requires that the local state estimates are independent of each other or their correlations are known, and it is generally very difficult to accurately calculate the correlation of local state estimates.

Images