Unmanned aerial vehicle ad-hoc network (UAVANET) has been paid more and more attention and applied in modern war, because it does not need the help of ground base station, flexible control and other characteristics. As an important part of its communication technology, the routing protocol of UAVANET has been put forward strict requirements. Therefore, a multi-attribute bidding model based on game theory is established to select routing nodes. Starting from node energy attribute, forwarding Angle and distance, the scoring function and revenue model of bidding nodes are designed. Finally, the best routing node is selected by comparing the quality attribute and bidding price of multiple bidding nodes. The simulation results show that the new node selection algorithm used in GPSR protocol can reduce transmission delay and increase packet delivery rate, and at the same time, it can ensure the energy balance and link stability of the network, and meet the requirements of UAV network for communication networking.