Distributed Guidance Of Unmanned Aerial Vehicle Swarms:Key Technologies And System Implementation

Unmanned Aerial Vehicle(UAV)swarms are showing great application prospect in many fields,such as communication coverage,disaster rescuing,military operations and so on.Compared with using a single UAV,utilizing a swarm has many advantages,which provides fault tolerance,task parallelism and function distribution,thus a swarm is capable of supporting complicated tasks,indicating its high research value.Guiding a UAV swarm is the procedure of moving the swarm to the target area according to the objective,which is the basis of performing various complicated tasks.So the guidance control of UAV swarms is a key direction of robotic research,which can be divided into centralized control and distributed control.Although centralized control is easier to be mathematically analyzed and has a stronger theoretical basis than distributed control,it is neither practical nor extensible in large-scale scenarios where communication is highly limited,which can be commonly found in many practical applications.Therefore research on distributed guidance of UAV swarms is of important theoretical and practical significance.In this paper,the guidance problem is divided into two parts,namely decide and act,based on the behavior model of robots.First,a centralized swarm path pre-planning for the UAV swarm is conducted according to the target.Then a distributed control is utilized to move the whole swarm.To support fixed-wing UAV swarms,we propose a distributed action control method for nonholonomic systems,which considers the formation control and group maneuverability simultaneously.The convergence of the control algorithm is proved and the stall speed restriction of fixed-wing UAVs is satisfied.Experiments in MATLAB are conducted to verify the correctness of the proposed control algorithm.We also set up a micro-rotor UAV hardware and software experimental environment,utilizing low-cost,highly integrated and high performance devices to build the hardware platform and open source Robot Operating System to build the software platform.Some additional functions like time synchronization are also provided to support and improve the efficiency of developing,debugging,deploying and experimenting with UAV swarm applications.With the theoretical research and the experimental environment proposed in this paper,a distributed guidance system for a UAV swarm is designed and implemented,with which some simulation experiments and physical experiments are conducted.The system consists of ground station nodes and UAV nodes,capable of controlling the whole flight process from the takeoff to the planning,the action and the landing.With the proposed control method,the guidance of the swarm is achieved utilizing local communications.We also develop human-machine interfaces and flight safety modules.The experimenting results show the proposed distributed guidance system for UAV swarms in this paper is practical and extensible,and verify the effectiveness of the proposed distributed control method.