Research On Autonomous Positioning Algorithm Of Leader-follower Micro-nano Satellite Cluster

With the development of satellite technology,individual satellites are unable to meet the needs of space missions.Micro-nano satellite cluster is currently widely used,providing many conveniences for people's lives.In order to make full use of the advantages of satellite cluster,satellite cluster needs to achieve autonomous localization.Fingerprint localization has the advantages of low cost and simple structure.Therefore,it is of great significance to extend it to the localization of the leader-follower micro-nano satellite cluster.This paper proposes an autonomous localization algorithm of leader-follower micro-nano satellite cluster based on location fingerprint.(1)Aiming at the problems that the fingerprint localization has high complexity and is greatly affected by time offset when it is directly applied to satellite cluster,a coarse-fine localization algorithm is proposed.In the offline stage,sparse radio maps are established through the propagation model.In the online phase,the follower satellite performs coarse localization according to the sparse radio map and determines the fine localization area Finally,the dense radio map of fine localization area is calculated and fine localization is performed.The proposed algorithm can effectively reduce the space and time complexity and can greatly reduce the impact of time offset.(2)Aiming at the low localization accuracy of the coarse-fine localization algorithm,this paper proposes a tracking algorithm based on two-step polynomial fitting.Firstly,the follower satellite fitting trajectory is determined by two-step polynomial fitting and the two-step fitting trajectory is compared with the localization results to obtain the trajectory consistency.Secondly,for the localization results that meet the tracking conditions,the localization results are tracked by establishing a tracking radio map.Finally,the Kalman filter algorithm is used for filtering.Experimental results show that the tracking algorithm can effectively track the follower satellite and can significantly improve the localization accuracy of satellite cluster on the basis of the coarse-fine localization.In summary,on the basis of fingerprint localization,we perform coarse-fine localization by establishing sparse radio map and dense radio map,which effectively reduces the algorithm complexity and can resist the impact of time offset.In addition,predicting and tracking the trajectory of follower satellites through two-step fitting and establishing tracking radio map can effectively improve the localization accuracy.