Collision-free Time-varying Formation Control Of Multiple Unmanned Surface Vehicles Subject To Unknown Input Gains

In recent years,with the development of complicated and large-scale surface operations in ocean space,the research trend of unmanned surface vehicle(USV)technology has been expanded from single USV to multi-USVs formation.Compared with single USV control,multiple USVs have higher work efficiency,better fault tolerance and stronger adaptability,and are widely used in civil and military fields such as environmental monitoring,search and rescue,resource detection and fleet replenishment.In this context,the control technology of multi-USVs is receiving unprecedented attention.How to realize the formation navigation of multi-USVs on the water surface and how to deal with the obstacle avoidance in critical situation has become the focus of the research on USVs.Based on the needs of practical problems,the distributed time-varying formation control problem of a class of underactuated autonomous surface vehicles under unknown input gain,model uncertainty and ocean disturbance is studied in this paper.Specific research work includes:Firstly,the formation control problem of underactuated USVs with unknown input gains,model uncertainties and ocean disturbances is investigated.A path-guided distributed time-varying formation controller with collision avoidance and connectivity preservation is designed.At the kinematics level,a distributed guidance control law based on the consensus approach,path-following design,artificial potential function and auxiliary variable method is presented.At the kinetic level,an adaptive kinetic control law is designed based on an indirect model reference adaptive control approach where a neural estimator is developed for identifying unknown input gains,model uncertainties and ocean disturbances.The adaptive kinetic control method based on the indirect reference model can ensure that the tracking error converges to a zero domain,so that the surge and yaw speed can achieve satisfactory tracking effect.The input-to-state stability of the closed-loop system is established via cascade theory.Finally,simulation results are provided to validate the effectiveness of the proposed time-varying formation control method for collision-free multiple underactuated USVs with unknown input gains.Secondly,a time-varying formation controller for underactuated USVs based on integral concurrent learning(ICL)method is designed to solve the problem of control of underactuated USVs with unknown velocity of neighbor USVs,unknown input gains,model uncertainties and ocean disturbances.At the kinematics level,an extended state observer(ESO)is introduced to estimate the velocity of neighbor USVs.At the kinetic level,an adaptive kinetic guidance law based on the integration parallel learning is proposed by using the ICL method to accurately identify the unknown input gains and the neural network to estimate the ocean disturbance and model uncertainties.Without continuous excitation,this method not only guarantees the convergence of tracking error,but also ensures the convergence of parameters,so that the estimated value of parameters is infinitely close to the real value,so as to achieve accurate estimation of unknown input gains.The input-to-state stability of the closed-loop system is established via cascade theory.Finally,simulation results are provided to validate the effectiveness of the proposed time-varying formation control method for collision-free multiple underactuated USVs based on ICL method.