Research On Cooperative Task Allocation Algorithm For Multi-underwater Robots

With the development of science and technology,the exploration of the ocean becomes more and more deeply and the underwater robot is an important tool for the exploration of the ocean.People hope robot can complete such tasks as undersea terrain exploration,resource exploration,and even military operations.These complex tasks are often difficult to complete by a single AUV or the efficiency is low,then the application of multi AUVs comes into the people's vision.And in the face of the complex task,how AUV perform the task to make full use of its effectiveness.So the multi AUVs tasks allocation problem becomes a main research direction.Ant colony algorithm is a swarm intelligence algorithm,with its unique parallel distributed computing,simple,positive feedback characteristics are widely used,but there are also some shortcomings,such as the algorithm is easy to fall into local optimum,stagnation and so on.In this thesis,we use the appropriate optimization measures to improve the ant colony algorithm,and the simulation results are verified.Particle swarm optimization algorithm is a kind of swarm intelligence algorithm,which is widely used to solve the problem of continuous with the characteristics of high search speed and high efficiency.But the local search ability of the algorithm is easy to fall into local optimum.In this thesis,we propose to improve the local search performance by adding a linear inertia weight which is different from the general weight value.The main research contents of this thesis are as follows:1.Select a centralized architecture with only one master AUV unit.And on the basis of the centralized architecture,the total path and the angle between the AUV and the target are selected as evaluation indexes,and the task allocation model is established.2.Analysis the existing problems of ant colony algorithm,Aiming at the existing problems in the ant colony algorithm,use the appropriate optimization method to improve the algorithm,and use the Matlab to carry on the simulation to verify.3.Research on particle swarm optimization algorithm.According to the problem that the algorithm has no better local search problem,the method of increasing the linear inertia weight value is put forward,.The particle swarm optimization algorithm has good global searching in the initial stage,and the local search performance of the algorithm is improved by reducing the weight value after narrowing the range.At last,the improved algorithm is also simulated by using Matlab.