Research On Finite-time Hunting For Multi-robot System

Compared with a single robot,multi-robot system has incomparable advantages,which has become a hot topic of research in recent years.Multi-robot hunting is one of the important issues,and it has significant research value in fishery,national defense and security fields.In view of the situation that the time limit of multi-robot hunting is seldom considered in the current research,this paper studies the problem of multi-robot finite-time hunting by forming formation method.The main contents are as follows:(1)In order to solve the problem of fairness of multi-robots hunting in a given time,an algorithm based on fair assignment of tasks is proposed.Firstly,the fairness coefficient is defined as an indicator to measure the fairness of task allocation,and every robot is assigned a task,which is a target point.Secondly,after the task assignment is completed,a periodic path planning is designed to enable the robot to move to the target point at a given time.Finally,the simulation results show that the difference of driving distance between different robots using this scheme is smaller than that without this scheme,which reflects fairness.(2)A new finite-time controller is proposed for finite-time tracking of a single wheeled robot.Firstly,the tracking error model is established by chain transformation of kinematics model,and the error system is decomposed into two subsystems.Then,based on the method of Lyapunov function and the homogeneous theory,the finite-time controller is designed according to the error system.(3)Aiming at the problem of finite-time cooperative hunting of multiple wheeled robots,a finite-time hunting method based on circular formation is proposed,considering the fixed directional communication topology of the robots.Firstly,a finite-time observer is designed to estimate the target states because some robots cannot obtain them.Secondly,based on the method of ring formation,the tracking target point is assigned to corresponding robot and the error systems are established.Finally,based on graph theory,Lyapunov function method and homogeneous theory,the distributed finite-time controller is designed according to the collaborative error,and the finite-time hunting task of multiple wheeled robots is realized.