A Time-evolving Topology Based Obstacle Avoidance Algorithm For Multi-UAV Formation

With the rapid development of the artificial intelligence technology,the application of UAV has gradually expanded and it become more civilized in plant protection,disaster relief,geological survey and military fied.However,due to the small scope of operation of a single UAV,limited field of view,low efficiency and otherfactors,making the demand of multi-UAV formation increased.The multi-UAV formation mission environment is a highly dynamic and complex environment,so the performance of the formation obstacle avoidance algorithm plays an important role in the process of multi-UAV formation of the whole task.In formation process,formation requires not only continuous follow-up changes of environment,but also topological relations among nodes.So the inplementation of the task in the actual environment brought some challenges for the formation of obstacle avoidance algorithm design.But in order to simplify the research conditions,many of the current study only assumed that the formation is carried out under fixed ideal communication conditions,and don't consider the constraints of communication conditions in the formation avoidance process.In this paper,the obstacle avoidance algorithm for UAV formation based on time-varying topology is developed for the unmanned aerial vehicle formation obstacle avoidance task in 3D dynamic environment.At first,the study considered the high dynamic environment in the actual flight,make a time-varying topology updated method to change the communication topology between each node in the formation process.It can avoid that the node loss situation which will happened under the fixed communication topology UAV formation.And then,this paper use the convex optimization theory and the artificial potential field theory to model the UAV formation,and established a distributed and centralized transmission method to reduce the communication overhead and time cost.And use the local informations on single node to make a global information to constraints formation to avoid obstacles with the whole nodes.That can avoid the formation separate situation which is because the node avoidance with itself and can ensuring the integrity of the UAV formation data link.Finally,the paper use the geometic principle designed the method to choose the middle point's location in the path,let the UAV formation can reach the destination without set the point in the path and achieve a autonomous formation obstacle avoidance algorithm in the dynamic environment.In this paper,we use the Matlab simulation platform to verify the algorithm,complete the simulation of the formation avoidance algorithm of multi-UAV in static environment and dynamic environment.Also analyzed the simulation results and compared the communication overhead and time cost with the general distributed UAV formation.The result shows that our not only can satisfied the dynamic environment,but also less on the communication overhead and less than 30% on the time cost in the whole cycle time of the algorithm when compared with the general distributed UAV formation algorithm.