This paper considers the problem of allocating heterogeneous unmanned aerial vehicles(UAVs) to varied types
of targets performing detection, attack and verification tasks. The amount of information gained through detection and
verification tasks is calculated by using the changes in entropy. The ability of UAVs performing attack tasks is modeled by
the probabilities of targets destroyed. The relationships between the tasks are also considered. The model of the distributed
task allocation for heterogeneous UAVs is established. And the distributed auction algorithm using limited communications
is developed to solve the task allocation problem optimally. Simulation results show the rationality of the model and the
effectiveness of the proposed method.