The Research Of Robot Path Planning Based On Genetic Algorithm

With the maturity of related fields such as computers,people have new requirements for their navigation technology such as intelligent robot.The complex diversity in their working environment makes the robot field more and more important in path planning.The genetic algorithm is an algorithm that finds the ideal solution through iterative iteration according to the law of biological evolution,which very effectively provides a optimization strategy in the field of path planning research.By using the high search efficiency and better scalability of genetic algorithm,this paper mainly uses the global optimization of genetic algorithm to reasonably plan the path of the robotIn this paper,we use genetic algorithm and elite strategy non-dominated sorting genetic algorithm to reasonably design the optimal path according to the single-objective and multi-objective problems that robots encounter in the working environment.Firstly,the basic theory of genetic algorithm and the process of stochastic iterative evolution in genetics are analyzed on the basis of describing the research status and development trend of robots.The contradictory multi-objective research shows the advantages of NSGAII in solving problemsSecondly,the environment of the robot is built by using grid method to preserve the environmental information and genetic algorithm to optimize the global characteristics.Finally,the process of the robot reaching the prescribed area is realized.In the past,the original algorithm used the strategy of simple random generation in the initial population,which would lead to too many infeasible paths affecting the search speed.In this paper,heuristic method is used to reduce the infeasibility of paths.In order to make the genetic algorithm not easy to occur local premature phenomenon,the new solutions obtained by crossover and mutation are based on the individual's fitness.Considering the randomness of mutation of the original algorithm,the neighborhood range search of the random points of the path is adopted to ensure the integrity of the path.On the other hand,population diversity is introduced to adjust the crossover and mutation probability adaptively to optimize the algorithm.Then,the simulated annealing algorithm is used to select the new generation of better individuals to retain to the next generation,and ultimately the optimal path can be found quickly.Finally,the objective functions of navigation distance,threat cost of environment and concealment performance of navigation are established on the optimization of robot's three-dimensional environment track.According to the randomness of the initial population of the original NSGA algorithm,the iteration time will be long.The chaotic idea is introduced to optimize the population,which avoids the monotony problem.On the other hand,the local search strategy is conducive to improving the performance of the algorithm.Finally,the optimal solution of each index function is found by the entropy method.