Research Of Multi-uavs Formation Control Technology

The Unmanned aerial vehicle(uav)in military reconnaissance and civil more and more widely.The single uav mission is relatively simple,sometimes more complex and difficult task depends on uavs work together.More uav formation can be expanded field of vision,increase reconnaissance and search scope,Therefore,formation flight control technology has become a hot topic for scholars to study.This paper mainly focuses on formation control,flight path planning and obstacle avoidance of multiple uav.Firstly,this paper proposes leader-follower method to solve the formation control problem of multiple uav,and studies the formation structure of multiple uav.Multiple uavs form a triangular formation in a two-dimensional plane.In the leader-follower method,one uav is set up as the leader,and the rest of the wingmen follow the leader.The leader determines the movement trajectory of the whole formation.In the formation,the expected wingman position information is calculated by the relevant information of the leader and transmitted to the wingman to make it move to the expected position in the formation.In the simulation experiment,through the position information of the leader and formation structure,the expected position information and the expected flying direction of the wingman in formation can be calculated.Secondly,in the uav flight path planning problem,this article put forward the path planning algorithm based on the NSGA-?.In order to be more practical,this paper simulates the terrain environment of uav flight and establishes a 3d terrain model.Due to the influence of flight path planning on concealment,path length and safety,NSGA-? algorithm is used to transform the uav path planning problem into a multi-objective optimization problem.In this paper,the multi-objective path planning problem is studied,and the path planning is optimized simultaneously on the basis of the minimum cost of the three objective functions of concealment,path length and safety.The simulation results show that the three objective functions have coevolved,the convergence speed is fast,the flight altitude change is small,and the flight concealment is maintained at low altitude,which verifies the effectiveness of the algorithm in flight path planning.Thirdly,in the obstacle avoidance problem of multi-uav formation,this article put forward an improved artificial potential field algorithm,which can provide sensitive collision avoidance capability for uav.Because the classical artificial potential field algorithm(APF)is easy to converge to the local minimum point,it produces the target unreachable phenomenon.To solve this problem,the potential field function of the classical APF algorithm is improved.In the original potential field function,the relative distance between the target and the uav is added.Simulation results show that the improved APF algorithm sloves the problem of unreachable targets in formation flight.In addition,in jitter problem,this paper proposes a dynamic step length design,The dynamic step size adjustment method can not only change the movement direction of the uav appropriately,but also reduce the step size when the resultant direction of the two steps changes obviously,so that the uav can avoid the shaking area gently.Finally,a simplified formation flight model based on kinematics constraints and uav dynamics was tested by using the leader-follower structure and the improved artificial potential field method.Experiments show that the proposed method improves the obstacle avoidance ability of multi-uav formation and the collision avoidance function of uav.The simulation results prove the effectiveness of the proposed method.