Research On Cooperative Searching And Hunting Method For Multiple Unmanned Surface Vehicles

With the development and exploration of the ocean,unmanned surface vessel(USV)has been widely used as a small,high-speed unmanned ship.Many unmanned surface vessels(USVs)are more efficient than single USV in performing maritime tasks.Therefore,it is an important direction for future USVs to study how to cooperate with multiple UAVs to accomplish maritime tasks.In this paper,the cooperative search method and target capture method of multi-USV system are deeply studied.The improved particle swarm optimization algorithm is applied to the search task of multi-USVs,and the target capture method based on neural oscillator is also studied.The main contents of this paper are as follows:Firstly,the individual architecture of USV and the group architecture of multi-USV system are established.The individual architecture divides the USV system into seven modules: command and control module,information acquisition module,navigation control module,energy management module,communication module and load management module.Distributed architecture is used to balance autonomous decision-making and cooperative motion of USV.At the same time,the three-degree-of-freedom kinematics model of the ship is established according to the motion characteristics of the unmanned aerial vehicle.Then,a multi-USV search method is proposed to search the target USV in the location environment.The search area is discretized into grid area.Then,the motion model,collision type and communication mode of the USV system are analyzed.Then,the update method of the moving target point and target point set of the USV is analyzed,so that the USV can select real-time and dynamically.Selecting and updating moving target points,then establishing fitness function based on artificial potential field method.Finally,the proposed algorithm is simulated.The simulation results show that the proposed search method can successfully search the target USV in three different search environments,and compare it with coverage search method and random search method.The simulation results show that the proposed multi-USV cooperative search method is effective.The cable method has high performance.Finally,a method of multi-USV target capture based on neural oscillator is proposed.The basic principle of extended Kalman filter is introduced.The virtual target points of the target USV are obtained by using extended Kalman filter.The radial distance and phase angle of each USV relative to the virtual target points are calculated,and the phase angle update common is established.The radius of the limit cycle is the radius of the limit cycle,and the phase update formula of the limit cycle is established to make the USV approach the virtual target point gradually and keep a distance with the target USV eventually.The speed update formula and heading angle update formula of USV are established to make USV adjust its speed and direction of motion in real time according to phase and radial distance.Finally,simulation experiments are carried out using the proposed capture method.The simulation results show that the multi-USV system has successfully achieved the target.Round up.