16 January 2023 in IEEE Transactions on Intelligent Vehicles
For civil applications of Unmanned Aerial Vehicles (UAVs) for search, and emergency, data transmission between UAVs and Ground Control Stations (GCS) is always suffered from long distance, undulating terrains or other interferences. Thus, relaying is a critical enhancing technology in wireless communications for better service quality. In this paper, a planning method for fixed-wing relay UAV (RU) serving to sequential tasks of UAV swarm is proposed. Firstly, considering characteristics of signal propagation, antenna, environment and UAV, the UAV-to-GCS and UAV-to-UAV communication models are designed. And the limits of communication are investigated, which are crucial issues to provide a guarantee for swarm mission. Meanwhile, based on the analysis of relay UAVs requirements during the spatial distribution of tasks, a mixed integer nonlinear programming problem (MINLP) is constructed to solve the approximate number of relay nodes and access locations. Then aiming for the minimum number of relay UAVs, a continuous time-sharing relay UAVs reward model is constructed, and the consensus-based bundle algorithm (CBBA) is adopted to obtain the maximum rewards, including the number of relay nodes, mission sequences and route planning. Finally, the numerical simulation for a typical scenario is conducted to show the performance of the proposed approach.
19 December 2023 in IEEE Transactions on Vehicular Technology
Communication relay is necessary for a swarm of collaborative Unmanned Aerial Vehicles (UAVs) executing missions in a large area, such as cooperative exploration, mapping tasks and so on, which assists for the quality of transmission in the swarm network. However, it remains a challenge to plan the mission of Relay UAVs (RUs) to enhance communication of the swarm considering the kinematics of UAVs and task sequences of a swarm. This paper proposes a two-layer relay planning framework for a fixed-wing UAV swarm consisting of coarse planning at the Ground Control Station (GCS) as well as refined trajectory planning onboard. The GCS planning layer evaluates the relay requirements in the swarm network to generate the initial deployment of RUs, aiming to minimize their numbers while maximizing utility. The β-weak connectivity model is adopted to evaluate the swarm network topology constructed from the Minimum Spanning Tree (MST). Meanwhile, a Mixed Integer Nonlinear Programming Problem (MINLP) is employed to determine the initial deployment of RUs including the number and accessing locations. Moreover, the onboard planning layer predicts the states of the Mission UAVs (MUs) connected to the RU using the Kalman filter. Based on these predictions, the trajectories of each RU are autonomously planned online to continuously guarantee stable network connectivity, which is achieved through Distributed Model Predictive Control (DMPC). Finally, numerical simulations are conducted to show the effectiveness of the proposed method.