Research On Robot Path Optimization Based On Evolutionary Algorithm

Path planning research is an important research direction of robot intelligent research.The main task of path planning research is to search for a collision-free path from the starting point to the end point in an environment containing obstacles,and to minimize the length of the path.In general,the path planning method can be divided into a local path planning method and a global path planning method.Generally speaking,the local path search method has the characteristics of simple calculation and easy implementation,but the algorithm has fallen into a local extreme point,not It is beneficial to the global optimization of the algorithm.The global path planning algorithm generally has the advantages of self-adaptation,self-learning and global convergence,and has certain redundancy and robustness.The representative of the two algorithms is the artificial potential field path planning method based on the potential field theory and the genetic algorithm path planning method that simulates the natural biological evolution rules.Genetic algorithms have the advantages of adaptability,robustness,global convergence and parallel computing in the application of optimization problems.However,there is a great deal of randomness and blindness in the search process of the problem.At the same time,there are problems such as poor local search capability and slow search speed.Therefore,according to the characteristics of the genetic algorithm,the artificial potential field algorithm is combined with the genetic algorithm,and the local search ability and the real-time characteristic of the artificial potential field algorithm are applied to the traditional genetic algorithm.,The improvement and optimization of the standard genetic algorithm can improve the local search performance of the genetic algorithm,accelerate the convergence speed,and improve the ability of the algorithm to solve the optimization problem.The improved algorithm in this paper combines the characteristics of the two algorithms and the particularity of the path planning problem.With the genetic algorithm as the framework,the artificial potential field algorithm is incorporated into the crossover operator in the genetic algorithm to improve the optimization performance of the crossover operator.,and the map information contained in the potential field map is applied to the search process of the genetic algorithm to improve the local search ability and global convergence efficiency of the genetic algorithm;and the drift operation of the path point is added to improve the efficiency of population initialization and mutation operation;The dynamic parameter setting makes the algorithm get a good balance in the search process,reduce the probability that the genetic algorithm gets into the local extremum,and make the algorithm get better global optimization ability.