A nonlinear three-step filtering method for Mars dynamic descent

Abstract

The invention discloses a nonlinear three-step filtering method for the Mars power descent stage, which comprises the following steps: 1, building a kinetics equation and a measurement equation of the discrete type nonlinear three-step method; 2, presetting the initial value; 3, performing state quantity filtering; 4, performing error filtering of a kinetics system; 5, error filtering of an unknown measurement system of a measurement system; 6, updating the correlation coefficient, and correcting the state estimation and the kinetics error estimation; 7, returning to the step 3 to go ahead till equaling to the time T corresponding to the deadline of the Mars power descent stage, namely, the landing time of the Mars lander, so as to complete the nonlinear three-step filtering method of the Mars power descent stage. The method fully considers the estimation problems about position and speed of a detector of a nonlinear non-Gaussian stochastic system under the condition of errors of the kinetics system and errors of the unknown measurement system in the measurement system during Mars actual power descent, and efficiently ensures the position and speed estimation of the detector of the Mars power descent stage.

Description

technical field [0001] The invention relates to a nonlinear three-step filtering method in the Mars power descent section. It belongs to the technical field of aerospace navigation. Background technique [0002] Mars is a planet that is closer to the earth. It has similarities with the earth in many aspects and has become one of the preferred target stars for human deep space exploration. So far, humans have conducted a total of 39 Mars explorations, 2 / 3 of which ended in failure. Obviously, due to the harsh environment of Mars and the immaturity of deep space exploration technology, human exploration of Mars is facing severe challenges. In Mars exploration, the most meaningful and difficult thing is the soft landing of the probe, and the entry, descent and landing process in the soft landing process are the key to the entire exploration mission. Of the 15 Mars landing missions that have been implemented, only 7 have landed successfully. From the moment the Mars rover le...

AI Technical Summary

Problems solved by technology

Although the unscented Kalman filter method can well apply the nonlinear Kalman filter, it is difficult to estimate and reduce the error of the dynamic system when the measurement means are limited and the measurement data is small, and relevant scholars have confirmed that the unscented Kalman filter is suitable Data reconstruction based on the state estimation of the Mars rover, so this method is not suitable for the Mars dynamic descent stage

Images