Multi-UAV task allocation and path planning in emergency rescue scenarios with uncertain requirements

UAVs have significant advantages in disaster relief material distribution, but the dual challenges of environmental dynamics and demand uncertainty restrict the coordination performance of multi-UAV. So, in this paper we propose a multi-UAV task planning method that combines a new robust optimization model and two improved intelligent algorithms. First, we establish a new robust model in which the uncertain set is constructed by the connection between the proportion of the affected population and the material demand. Then, we seek the minimum allocation cost under the maximum demand fluctuation in order to counter the high uncertainty of demand. An improved genetic algorithm is provided to solve the robust optimization model, and it has obvious advantages over other algorithms, including the Gurobi solver, in terms of computational efficiency and rationality. Second, we construct a multi-UAV path planning model, which takes into account dynamic constraints and collision avoidance. An improved artificial fish swarm algorithm is adopted to solve it. The experimental results show that this method has strong adaptability and superiority in emergency rescue scenarios.