Research On Unmanned Surface Vehicles Swarm Formation Control Based On Flocking Algorithm

The Unmanned Surface Vehicle(USV)swarm formation consists of several USVs,which interact with each other and give target instructions through central information to accomplish various tasks.USV flocking has long operation time,low cost,high operational efficiency,and can save a lot of manpower,instead of human to carry out dangerous operations,and can complete a variety of tasks such as detection,search and rescue,patrol,surveillance,etc.,and has high application value in both civil and military.This paper focuses on the USV flocking formation control method based on swarming algorithm,and the main research contents are as follows.1.A USV-Agent tracking control mathematical model for flocking control is proposed and established.The conventional flocking control is based on the prime point model,but the prime point model is not sufficient to describe the USV object with obvious dynamics characteristics and serve as the model basis for formation control.To solve this problem,a target tracking control strategy based on LOS guidance algorithm and adaptive sliding mode control law is designed to control the USV heading and longitudinal direction separately,and the mathematical model of unmanned USV-Agent tracking control is obtained and validated by simulation.2.In response to the situation that the speed of the virtual navigator in the classical Olfati-Saber flocking algorithm is constant and does not match with the actual situation,a multi-USV flocking formation control strategy with a variable-speed virtual navigator is designed.Considering that the virtual navigator is affected by external forces with time-varying speed,a virtual pilot compensation term is added to the classical Olfati-Saber flocking algorithm based on the USV-Agent tracking control model,so that the speed of the agent can converge to the same speed as the virtual pilot and reach a stable flocking.The application of the flocking control algorithm with variable-speed virtual navigator to the swarming control of USV cluster formations in the presence of obstacles is further investigated.The effectiveness of the algorithm is verified by simulation experiments.3.To address the problem of possible swarming breakup in the presence of non-informative agents,a flocking swarm formation control strategy for USVs based on the idea of local tethering is studied and designed.The local tethering algorithm is applied to control the class Ⅱ non-informative agents,so that the non-informative agents can be connected to the network through their nearest informative agents to form a topological network when the topology is not connected;in order to make the information pass to the whole network,the local tethering control term is further introduced to reach the control of the whole network under the premise of flocking control,so that the agents reach flocking.Finally,the effectiveness of the algorithm is verified by simulation.