A distributed task allocation method for heterogeneous UAVs in dynamic and communication-constrained environments

This paper considers the multiple unmanned aerial vehicle (UAV) task allocation problem in dynamic and communication-constrained environments, where limited resources, non-ideal communication factors, deadlines and newly added tasks are considered. Given the complex constraints, the task allocation problem is modeled mathematically. The proposed method builds upon the performance impact (PI) algorithm, extending it with an improved task inclusion phase, a novel conflict resolution phase and a systematic way of clustering while preserving the robust convergence property. In addition, a coalition formation method is presented to cope with the newly added tasks or the tasks that cannot be completed by a single UAV after repeating the task inclusion phase and conflict resolution phase. Numerical simulations are constructed to illustrate the procedure and superiority of the proposed method compared with the state-of-the-art distributed task allocation algorithms. The proposed method is not only effective in increasing the number of tasks with resource requirements satisfied but also efficient in saving the number of iterations and communication events to converge.