In resource-constrained underground mines, traditional localization techniques are usually unable to meet people’s needs for localization accuracy. Due to the scarcity and limitation of localization resources, the nodes to be localized are required to adopt effective node selection as well as resource allocation strategies to ensure that the nodes have sufficiently high localization accuracy. However, nodes with high localization accuracy in the network usually do not want to sacrifice their information to help other nodes, so we adopt monetary incentives to improve the motivation of nodes in collaborative localization. In this paper, we propose a strategy based on competitive proportional auction for resource allocation, and introduce the integrity coefficient of nodes to complete a feedback mechanism, with the goal of minimizing the squared bounds of nodes’ location errors, and introduce currency as an incentive indicator to further reduce the global resource allocation loss. The results show that the method can effectively improve the positioning accuracy, reduce the energy consumption, and exhibit good robustness in the dynamic network environment.