Research On Cooperative Task Allocation Method Of Multiple Unmanned Surface Vehicle Under Uncertain Conditions

Efficient cooperative task allocation method is the basis for realizing multi-USV cooperative mission planning,and it is a hot and difficult problem in the application field of multi-USV mission planning.Uncertain factors such as the complex and changeable sea environment and the incompleteness of perceptual information pose new technical challenges to the coordinated task assignment of multiple USVs under uncertain conditions.Aiming at the task assignment problem of multiple unmanned boat formations under uncertain conditions,this thesis aims at maximizing the benefit of task assignment and minimizing the cost,and adopts an improved wolf pack algorithm to assign tasks.The specific work is described as follows:(1)For the problem of maritime target threat assessment under uncertain conditions,a target threat assessment system is established for the interval information sensed by the unmanned boat system,and a method of interval factor TOPSIS(Approximating Ideal Solution)-grey correlation analysis is proposed to assess the target threat degree.(2)For the task assignment problem of multiple unmanned surface vehicle formations under uncertain conditions,the assessed target threat degree is added to the task assignment model,and an improved task assignment method of the binary matrix wolf pack algorithm is proposed.The improved strategy is to introduce the update ideas of the local leader phase and global leader phase in the spider monkey optimization algorithm into the running and wandering process of the wolf pack algorithm,so that there is information interaction between the wolf packs.The local and global optimization-seeking ability of the algorithm is enhanced,and the convergence speed is accelerated.(3)In the simulation experiment,the comparison experiment with the traditional TOPSIS and gray correlation analysis method verifies the rationality of the multi-USV cooperative target threat assessment system and the effectiveness of the TOPSIS-grey correlation analysis method based on interval factors.Through comparative experiments with other population intelligence algorithms,the feasibility of the task allocation method based on the improved binary wolf pack algorithm is verified.Through the pre-allocation and redistribution experiments of multi-USV cooperative tasks,the ability of the multi-USV system to deal with emergencies when performing tasks is verified.