Research On Autonomous Anti-collision System Of UAV Based On Formation Reconstruction

With the mature of unmanned aerial vehicle(UAV)related technology,the development prospects and potential application scenarios of UAVs are also recognized by more and more people.Among them,UAV formation is an important development direction in the field of UAVs in the future,it can overcome the limitations of a single UAV in terms of load,endurance and mission types.Anti-collision is a basic problem that UAV formation needs to solve in its application scenarios.Therefore,research on autonomous anti-collision of UAV formation has very important practical significance.Motion planning and obstacle avoidance are the most important components of an anti-collision system.The ultimate goal of motion planning is to allow each UAV in the formation to reach the end without collision,and obstacle avoidance allows UAVs to respond reasonably to obstacles during the flight.This paper proposes an UAV autonomous collision avoidance system based on formation reconstruction.Although the pre-generated reference path is optimal when the obstacle position is known,it does not consider the occurrence of unknown obstacles in the actual marching process of the formation.Based on formation reconstruction,this paper proposes a UAV autonomous anti-collision system that combines the characteristics of multi-rotor UAV to avoid unknown obstacles.It can generate collision-free trajectories for each multirotor UAV in the formation in real time without considering unknown obstacles in the reference path.First,through the UAV screening algorithm proposed in this paper,the UAVs within the formation that are most likely to collide with obstacles are screened out,and the proposed obstacle avoidance strategy based on formation reconstruction is implemented.After that,the UAV will move to the latest waypoint without collision under the guidance of an improved artificial potential field.Finally,after the obstacle avoidance is completed,all UAVs will resume their original formation and return to the pre-generated reference path.This method enables the UAV formation to deal with static unknown obstacles while flying along a pre-generated reference path,and finally reach the destination without collision.The simulation results verify the feasibility of this strategy.This article mainly carried out the following three innovations:(1)A screening method combining the current state of UAV,the influence range of obstacles and the movement trend is proposed to screen UAV in formation.In addition,priority should be given to those UAVs with the greatest possibility of collision with obstacles,so as to reduce the computational load of the system and thus shorten the response time.(2)This paper proposes a distributed UAV formation collaborative obstacle avoidance method based on formation reconstruction.By changing the formation shape,the formation along the reference path can avoid unknown obstacles.The artificial potential field method solves the unreachable point and local minimum problems existing in the traditional artificial potential field and makes the artificial potential field method adapt to the three-dimensional environment.(3)By assigning a preset height to the UAV that perform obstacle avoidance operations,the collision of UAVs in the formation is effectively avoided,and the collision within the UAV formation is not affected by the number of UAVs in the system,reduce the communication link between the UAV,and thus have better scalability.