Design Of Distributed Pattern Formation Algorithm Of UAV Swarm

UAV(unmanned aerial vehicle)formations are widely used in many fields,such as geographic exploration,military reconnaissance,and so on.The research results of distributed formation of UAVs are still immature,and there are many practical constraints in practical applications,including collision avoidance between UAVs,limited communication distance,limited endurance,and limited computing power of onboard small computers,which lead to the formation algorithm of the UAV swarm designed under distributed conditions cannot adapt to many formation tasks.Aiming at the many constraints of the formation of UAV swarm in the real world and the practical need to design corresponding formation schemes for different application scenarios,this paper studies the constraints of the limited communication distance of the UAV swarm and the collision avoidance constraints of the formation process of the UAV swarm.The UAV swarm completes the formation of the specified pattern at the lowest possible cost under these two constraints.By analyzing the four fields including UAV swarm formation,robot formation,potential energy field model and rigid body model,the UAV swarm distributed pattern formation algorithm is designed.The specific research content is as follows:The convergence-separation match algorithm for UAV swarm is proposed to solve the distributed pattern formation problem of UAV swarm.In this paper,by constructing a fully connected set of UAV swarm for communication,each UAV can obtain information about all UAVs in the UAV swarm through multi-hop communication and depth-first traversal.By constructing a rotatable coordinate axis and using the best matching method to compute,the UAV obtains a UAV swarm formation scheme with collision avoidance as well as the flying cost of the UAV swarm as small as possible,which complete the task of constructing the target pattern of UAV swarm.Through simulation experiments,the algorithm is analyzed,and the correctness of the convergence-separation match algorithm for UAV swarm is verified.Through comparative experiments with the centralized algorithm,the algorithm designed in this paper has a smaller total communication distance in the formation process.Through comparative experiments with Direct algorithm and the comparative algorithm designed,the movement cost of the algorithm designed in this paper is relatively small,which verifies the effectiveness of the convergence-separation match algorithm for UAV swarm.Through extended experiments,the algorithm designedin this paper can be extended to formations with irregular patterns.A distributed pattern formation algorithm for UAV swarm based on the combination of imitate-rigid body and potential energy field is proposed,which solves the problem of distributed pattern formation for UAV swarm under the constraints of limited communication distance and collision avoidance.In this paper,the concept of imitate-rigid body proposed is combined with the characteristics of the UAV to obtain its own coordinate position in real time to obtain a collision avoidance formation scheme to complete the task of constructing a target pattern for the UAV swarm.Aiming at the local optimal defect of the potential energy field method and the defect of formation failure under the constraints of the limited communication distance of the UAV,the corresponding collision avoidance scheme of the imitate-rigid body was constructed by the potential energy field method,which eliminated the limitation of the limited communication distance of the pure potential energy field method.It also reduces the formation cost of the UAV swarm formation into a specified pattern.Through simulation experiments,the correctness of the algorithm is verified,and the algorithm can compensate for the defect of formation failure caused by the pure potential energy field method due to limited communication distance under distributed conditions.Through comparison experiments with the UAV swarm distributed formation algorithm of pure potential energy field method as well as Direct algorithm,the movement cost of the algorithm designed into a specified pattern is smaller,which verifies the effectiveness of the algorithm.