Abstract: ‍ ‍Large-scale unmanned Aerial Vehicle （UAV） swarms rely on collaborative awareness， information exchanging， and collaboration to complete complex tasks. With high intelligence and autonomy， UAV swarm has gradually become the future development trend of UAV technology. At present， large scale UAV swarm is closely combined with high-precision positioning technology. UAVs rely on high-precision relative positioning to maintain cluster stability and avoid collisions， and the entire UAV swarm relies on global navigation and positioning to achieve target guidance. In order to reduce costs， large scale UAV swarms generally use micro-electro-mechanical system （MEMS） inertial navigation systems and satellite navigation systems. However， in the existing main navigation positioning methods， there are serious cumulative errors in MEMS inertial navigation， and satellite navigation in complex electromagnetic environments is easily disturbed， both of which are difficult to achieve high-precision relative positioning and long-term global positioning. To address this issue， this paper proposes a fusion positioning method that relies on communication ranging technology to correct the error of inertial navigation and achieve high-precision relative distance measurement within UAV swarms， so as to provide high-precision navigation and positioning services for UAV swarms to perform tasks under satellite navigation refusal conditions. It can not only improve the relative positioning accuracy within the UAV cluster， but also improve the global positioning accuracy of the entire UAV cluster. In this paper， an inertial navigation positioning model and a communication ranging signal model are first established respectively， then a fusion localization problem based on the Maximum A Posteriori （MAP） method is constructed， and the optimal estimation problem of fusion positioning is constructed by the fusion of communication signal and inertial guide data. Secondly， based on the independence of inertial navigation and communication ranging， this paper derives the ranging accuracy from the communication signal delay estimation， and finally obtains the Cramér-Rao Lower Bound （CRLB） of the positioning problem. Then， this paper proposes a full information fusion localization method based on Gaussian Newton optimization to solve the fusion localization MAP problem， which iteratively converges to the optimal estimated position with inertial guided localization as the starting point. In addition， for the special scenario that the ranging accuracy is much higher than the inertial navigation positioning accuracy in long-distance flight， this paper proposes a high-precision ranging auxiliary positioning method based on the Scaling-by-Majorizing-A-Complicated-Function （SMACOF） algorithm， which realizes the optimal position estimation by establishing high-precision relative coordinates and optimal transformation parameter estimation. Finally， numerical simulation verifies that the proposed method can reduce the fused positioning standard deviation toof the inertial guided positioning standard deviation， and can gradually achieve CRLB performance. For a swarm of 100 UAVs， the method proposed in this paper can improve the accuracy of inertial guidance positioning by an order of magnitude. This paper utilizes UAV swarm quantity and communication ranging technology. On the basis of low-precision MEMS inertial guidance， it greatly improves the navigation and positioning performance of UAV swarms， enhances the overall positioning accuracy and relative positioning accuracy of unmanned aerial vehicle swarms， and effectively supports UAV swarms to perform long-distance and long-endurance tasks.