Unmanned Surface Vehicle Path Planning Based On Immune Genetic Algorithm

With the development of science and technology,many new unmanned products,such as unmanned aerial vehicles,unmanned surface vehicles,unmanned submarines and unmanned supermarkets have emerged.The emergence of unmanned surface vehicles provides a new style for offshore operations.In the non-military aspect,the unmanned surface vehicle can complete the tasks of water quality monitoring,patrolling and marine survey efficiently and safely.Path planning of unmanned surface vehicles is the core of unmanned surface vehicle control technology and global path planning is the basis of path planning.It relates to the maneuverability,economy,and security of unmanned surface vehicles.Therefore,this thesis studies of path planning unmanned surface vehicle based on immune genetic algorithm.Firstly,this thesis introduces the background and significance of the research of unmanned surface vehicles,and the status of unmanned surface vehicles at home and abroad.Then,introduced a variety of commonly used path planning methods,such as traditional methods,graphics methods,intelligent bionic methods and so on.This thesis introduced the principle of genetic algorithm in detail,including the basic concept of genetic algorithm,implementation process,advantages and disadvantages.For the poor search ability and slow convergence of genetic algorithm,a method based on elitism retention strategy is proposed.Aiming at the poor diversity of genetic algorithms,a path planning method based on immune genetic algorithm was proposed.The concepts of affinity,similarity,and concentration in immunology were applied to genetic algorithms and new antibody concentration definitions were used.This thesis uses the grid method to model the environment and preprocesses the depth information and width information of the unmanned surface vehicle in the environment.At the same time,Cartesian coordinates and serial numbers are used to encode the grid.The elitist retention strategy and the immunization operator are used,and the path search speed and path diversity are taken into account.And introduce insertion and deletion operator optimization path coding.Simulate under the platform of Visual Studio 2010,compare the performance traditional genetic algorithm,the genetic algorithm under the retention strategy,the immune genetic algorithm,and the global path planning based on elitism and immune genetic algorithm.It is concluded that the EIGA has better convergence speed and accuracy.