Research On Path Planning For Mobile Robot Based On Evolutionary Algorithms

Path planning is an important issue for mobile robot, which makes mobile robot to achieve destination according to set strategy. Path planning also a complex problem that not only guarantees a collision-free path with minimum traveling distance but also requires smoothness and clearances .Researchers have done a lot of works about Path Planning.As an intelligent search and optimization method, Evolutionary Algorithm (EA) has a good future with a very variety of applications.This paper is mainly research mobile robot's path planning by means of evolutionary algorithms. according to the characteristic of path planning problem, every component of the algorithms are analyzed carefully, including environment representation, chromosome representation, path evaluation, genetic operators design and GA parameters selection. after a comment on the advantages and disadvantages of the traditional optimization algorithms and the modern intelligent optimization algorithms when applied to the path planning of mobile robot, A new hybrid genetic algorithm is presented by use of genetic algorithm to optimize the valid path generated by chaos variation method. Combing the inversion property of the genetic algorithm with the ergodic property of the chaos optimization method,the path planning method based on a chaos genetic algorithm for mobliel robot can improve the local search ability and search accuracy of genetic algorithm ,and then effectively avoid the common defect of early convergence when using a simple genetic algorithm in path planning problems. A new kind of EAs has been proposed Estimation of Distribution Algorithms(EDAs). The algorithms achieve a good balance between good convergence and diversity. The two algorithms don't learn the complex structure information between variables on estimating probability distribution in evolution because of adopting simple probability model .So their computability complexity is low and speed is quick.Experiment is important for the study of path planning to validate the performances of algorithms and to make comparisons. Several path planning experiments are designed to prove the effectiveness as well as the efficiency of different algorithms.