Abstract: In heterogeneous multi-drone intelligent nest systems, the cooperative operation of large-scale UAVs can effectively avoid the operation interruption caused by UAV failure in a single nest system, which has a broad application prospect. However, existing studies mainly focus on small-scale scheduling problems, while giving insufficient consideration to scheduling large-scale charging tasks. In order to solve the scheduling optimization problem of heterogeneous multi-drone intelligent nests, multiple factors such as UAV path planning, conflict avoidance, and task allocation must be considered comprehensively. In this paper, a UAV route allocation model based on multi-computer nests is proposed. With task priority considered, an improved genetic algorithm for task allocation is designed. The effectiveness of the genetic algorithm in large-scale UAV route assignment is further explored by testing the algorithm under multiple environmental constraints. The results show that the method can adapt to changes in tasks, environment, and resources, and it demonstrates a good task allocation effect in large-scale heterogeneous multi-drone intelligent nest environments.