Research On Cluster Motion Control Of Intelligent Unmanned Surface Vehicle

With increasing importance being attached to the need for more extensive exploration and coastal defense system of the oceans,the demand for marine autonomous observation equipment is also increasing.The Unmanned Surface Vehicle(USV)is an important marine autonomous observation device with small size and strong environmental adaptability.The problem of regional coverage and task assignment requires both USV to have a high degree of autonomy and strong synergy.However,USV has lower autonomy than other unmanned carrier systems.This paper enhances the ability of USV to address area coverage and task allocation problems by simultaneously coordinating multiple USVs,enabling USV to have the ability to make autonomous decisions.This thesis mainly takes researches on the area coverage,obstacle avoidance path planning,as well as task allocation.The main contributions are as follows:In this paper,the distributed antiflocking algorithm inspired by the solitary behavior of some animals is used to solve the problem of area coverage using USVs on sea surface.Firstly,the problem description and system modeling of the USV implementation area coverage problem are studied.The distributed anti-flocking algorithm is studied to determine the repulsive pairwise potential function and the control input of USV.Secondly,in the process of USV information transmission,a simplified version of distributed information maps including sensing history is constructed.We also determine the communication rules between the USVs to update the information map.Then,the maximum profit target that satisfies the USV minimum turning radius and the maximum turning angle constraint is introduced to effectively obtain the coordinates of the USV which can maximize profit target at the next moment.Finally,the effectiveness of the proposed algorithm is verified by simulation experiments.For the problem of USV obstacle avoidance path planning in complex environments with static and dynamic obstacles,the distributed anti-flocking algorithm is developed for obstacle avoidance and path planning based on path prediction.Firstly,the selfishness term and repulsion potential functions of the anti-flocking algorithm with obstacle avoidance ability are improved,so that it can quickly track the target point and avoid the situation that the target cannot be reached.Secondly,the collision pre-judgment is carried out according to the motion state and the trend of the obstacle.Based on the result of the pre-judgment,the USV determines the path by making an early left or right turn.Finally,the simulation results is verified and the simulation experiments show that the proposed algorithm makes the system exhibits higher scalability and adaptability to environmental changes.For the task allocation problem of heterogeneous USV,a distributed sequential single-item auction scheme that takes the UAV limitations into account is studied.We also provide a systematic procedure for the auction process,so that USVs enable to make autonomous decisions when searching and completing tasks in an initially unknown environment.Depending on the decisions obtained from neighbors,an agent decides to auction or forfeit the target.The bidding and auction algorithms is introduced to obtained a conflict-free and feasible task sequence.We introduce a new bidding algorithm for USVs operating with local knowledge.This algorithm is designed for USVs operating with partial knowledge of tasks and USVs.It uses information of heterogeneous expertise to choose whether a USV should stay and complete tasks it knows about,or roam to find tasks that it is more suited for.Finally,the effectiveness of the designed algorithm is verified by simulation experiments.