Optimization study of unmanned surface craft formation control by fusing BSM and ANN

Abstract

With the continuous development of intelligent technology, the application of surface unmanned vessels (USVs) in the field of ocean engineering is becoming more and more widespread, and the cooperative cooperation of multiple USVs can improve the efficiency of task execution and expand the operation coverage, which has a broad application prospect and practical value. In this paper, the kinematic model of USVs formation under three degrees of freedom of longitudinal oscillation, transverse oscillation and bow rocking is established, and the dynamics model is built based on the internal structure of USVs, which lays the foundation for the study of the formation control and collision avoidance problems of USVs. Subsequently, the pilot-follower model is selected as the basis of the formation, the congestion control strategy based on the basic information model (BSM) is proposed to effectively alleviate the congestion of the fleet communication network, and the excellent approximation capability of the ANN-based RBF neural network is utilized to design the adaptive neural network controller. Finally, the event-triggered mechanism is combined with the adaptive dynamic planning algorithm to propose an adaptive optimal control strategy based on BSM event triggering, and at the same time, the neural network weight updating rules determined by the event triggering conditions are set to realize the optimization of unmanned surface boat formation control. Simulation experiments show that the unmanned boat can fulfill the formation task well, avoiding obstacles, maintaining formation, and maintaining target consistency. The proposed control method can significantly reduce the computational burden of the controller, ensuring the stability of the USV cluster system and alleviating the excessive utilization of system communication resources.