In this paper, a hierarchical coalition approach is proposed. The proposed approach is conducted under a hierarchical framework, which is composed of individual robots in the bottom layer and managers in higher layers. A bottom-up resources vector updating process for each manager is executed. Next, a top-down resources comparison process between the task and the managers is used to generate candidate robot coalitions. Specifically, different managers may be combined when necessary to meet the resources requirement of some complex tasks. Furthermore, a matching degree between the task and candidate robot coalition is utilized for an optimized coalition selection. The effectiveness of the proposed approach is verified by simulations.