Aggregation and Adjustment mechanisms for disaster relief task allocation with uneven distribution

The use of unmanned aerial vehicles (UAVs) to perform disaster relief task is becoming more and more popular. Generally, UAVs for rescuing are required to be heterogeneous, since the types of rescue missions are diverse. There are many researches on the task allocation of heterogeneous UAVs, in which they setup tasks to be randomly distributed when simulation. However, in most cases the distribution of tasks is regular rather than random. Therefore, we setup a set of special scenarios to simulate real disaster relief tasks, where most tasks concentrate in a few small areas and others are distributed randomly. LAL algorithm is an appropriate alternative to deal with assignment problems in scenarios where tasks are completely randomly distributed, but with a bit poor performance in the scenarios above mentioned. In order to obtain better results, this paper proposes three algorithms based on LAL algorithm in such scenarios. They are the Aggregation-LAL (A-LAL) using aggregation mechanism, Dynamic Adjustment-LAL (DA-LAL) using dynamic adjustment mechanism and Aggregation and Adjustment (AA-LAL) algorithms.Through the experiments in the NetLogo simulation environment, the proposed algorithms are evaluated. The result shows improvements, which are average 5% obtained by AA-LAL, of task allocation in our special scenarios compared with existing algorithms. Note: The project funding information is updated.