Graph-based modeling and non-decoupling solution for distributed cooperative mission planning

This paper addresses the problems of autonomous task assignment and path planning for a fleet of heterogeneous unmanned aerial vehicles in cooperative mission. In previous work many algorithms almost run on centralized architecture and handle the task assignment decoupling with the path planning. This may result in poor solutions. Therefore, this paper investigates a novel integrated solution for UAV to perform multiple consecutive tasks cooperatively on multiple ground targets based on distributed planning architecture. In a given scenario, the heterogeneous vehicles have different capabilities, kinematic constraints, and fuel constraints. Furthermore, the task has other constraints, such as task execution orders, UAV conflict free constraints, etc. This paper presents details of the non-decoupling solution which produces optimal assignment and trajectories for several given scenarios. The performance of the algorithm is compared to that of some previous methods in real-time simulation environment. The simulations results show the viability of the non-decoupling approach, and the non-decoupling solution has an advantage over hierarchical algorithms, and the distributed architecture improves the operation efficiency of the algorithm and the robustness of the UAV.