The rational layout of emergency facilities is an important guarantee for efficiency, fairness and stability of resource supply in post-disaster. Aiming at the uncertainty of sudden natural disasters, we study the robust optimization of hierarchical collaborative location of emergency facilities based on multiple coverage. Firstly, a multiple coverage level function is formulated based on the hierarchical facility location, and a bi-objective model of hierarchical collaborative location of emergency facilities is presented with the optimization objectives of minimum coverage level and expected total cost. Then, two kinds of robust optimization models are developed by applying the cardinal uncertainty set and the p-robust method to explore the influence of uncertain demand and random scenario on the facility layout. Finally, an empirical analysis on the location of seed reserve in Hunan province is carried out to validate the proposed models. The results show that there is a clear advantage to the layout solved by the collaborative location model over the traditional layout. The robust optimization models can effectively deal with the demand under the uncertainty and random scenario. The collaborative layout of facilities is affected significantly by the risk preference and budget level of decision-makers, so it is necessary to make a comprehensive trade-off between these.