Walker constellation orbit perturbation compensation method based on multi-agent reinforcement learning algorithm

Abstract

The invention discloses a Walker constellation orbit perturbation compensation method based on a multi-agent reinforcement learning algorithm. The method comprises the steps of A, initializing a state set S, an action set A and a Q value table of the reinforcement learning algorithm; B, designing an action a according to the current state and the sharing strategy, namely a sharing Q value table, obtaining current return r according to the action a, and renewing the Q value table; C, judging whether or not the current Q value conforms to the sharing strategy of the same orbit surface, namely the sharing Q value table, and if not, executing the step D; D, conducting calculating to judge whether a current geometric dilution precision (GDOP) value meets a design requirement or not, and if not, executing the step E; E, conducting looping execution on the step B to the step D till a GDOP evaluation model meets a long-term and stable running condition for a satellite constellation. According to the walker constellation orbit perturbation compensation method based on the multi-agent reinforcement learning algorithm, the instantaneity of conducting bias compensation on a normal orbit and the reliability of guaranteeing of long-time service of the satellite constellation can be improved.

Description

technical field [0001] The present invention relates to the compensation technology for the orbital perturbation of the Walker constellation, in particular to a compensation method for the orbital perturbation of the Walker constellation based on a multi-agent enhanced learning algorithm. Background technique [0002] Each orbital plane of the Walker constellation is evenly distributed, and the satellites in the orbital plane are also uniformly distributed. The satellite orbit is a circular orbit, and the Walker constellation is usually represented by N / F / P. Where N represents the total number of satellites in the constellation, F represents the orbital plane in the constellation, and P represents the phase modulation factor, that is, the number of satellites in each orbital plane is: m=N / F. A parameter to measure the performance of the constellation is Geometric Dilution of Precision (GDOP). The larger the GDOP value of the constellation, the angle from the ground receiver...

AI Technical Summary

Problems solved by technology

The impact of perturbation on the satellite orbit is mainly due to the drift of the right ascension of the ascending node and the drift of the track angle. All satellites in the satellite constellation will have a decline in the performance of the satellite constellation due to the drift of the right ascension of the ascending node and the track angle

[0004] Therefore, it is necessary to seek a reliable, low-energy orbit perturbation compensation method to solve the problem of satellite constellation performance degradation and realize long-term in-orbit service of satellite constellations

Images