Multi-agents Formation Control Based On Shape Theory And Flocking

The formation control of multiple Unmanned Aerial Vehicles(UAV)has played an important role in applications such as environmental awareness,cluster confrontation,search and rescue,etc.It is the basis of multiple agents formation navigation.Formation control mainly includes two research contents: formation shape control and formation maneuver control.In terms of shape formation,the current research is mainly focused on formation control of a group of formations,but this is not suitable for multi-task scenarios;in terms of formation maneuver control,most domestic and foreign scholars use a variety of methods to define formations.Realize the mobile control of the formation,the invariance of the definition formation method is weak.In order to make the formation adapt to multi-target or multi-task scenarios,and to improve the invariance of the definition of formation methods,this paper takes UAVs as the research object,mainly to study formation shape control and formation maneuver control:In terms of formation shape control,a group of formations cannot adapt to multi-objective or multi-task situations.This paper proposes an algorithm that can independently control the shape of multiple groups.The shape theory is used to define the formation shape,which transforms the formation shape control problem of the formation into a non-smooth convex optimization problem,and uses a recurrent neural network to solve the optimization problem.The algorithm can independently control the size,shape,direction and restricted area of multiple groups of formations;using the hungarian algorithm to recalculate the matching relationship between the initial position of the UAVs and the target position,reducing the total path length of the formation of multiple UAVs by about 45%.Two simulation examples are given to verify the effectiveness of the algorithm proposed in this chapter.For the formation maneuver control,a UAVs formation maneuver control algorithm based on shape theory and flocking algorithm is proposed.Use occupancy grid map to build a workspace for formations,set up restricted areas for formations based on the optimal path of the free grids.And the formation formation control algorithm based on shape theory is used to calculate the target formation in each restricted area that satisfies the preset constraint conditions.The flocking algorithm is used to drive the formation to track the time-varying target formation to complete the rotation,translation and scaling movements.Finally,the feasibility of the algorithm proposed in this chapter is verified by simulation.The degree of formation deformation and formation following the reference path are used to measure the performance of the algorithm,and the results show that the algorithm can move from the initial position to the target position while maintaining the formation.Based on the Robot Operating System(ROS),a Hardware-in-Loop Simulation Platform(HILSP)for multi-UAV formation was built,and the formation shape control algorithm was verified using this platform.