Real-time task reallocation for large-scale UAV team in dynamic environment

Abstract

To solve the problem of cooperative task reallocation for large-scale unmanned aerial vehicle (UAV) team under complex dynamic environment, a method based on man in-loop consensus auction principle is proposed. Firstly, Delaunay diagram is used for environment modeling. Then, the shortest paths of UAVs to targets are obtained by using Warshall-Floyd algorithm, thus task allocation and path planning are decoupled. Finally, with man in-loop and consensus-based auction approach, the task reallocation under several sudden changes of environment is achieved. Simulation results show that, the proposed method can provide a feasible task reallocation in 1.2s for a large-scale UAV team, which proves the real-time performance of this method.