Robot Path Planning Based On Ant Colony Algorithm And Simulated Annealing Genetic Algorithm

Path planning is a hot spot in the research of mobile robot navigation.2D path planning technology is mature,and has been successfully applied in engineering.In view of the robot work space being a 3D environment,path planning in 3D environment is highly necessary.The research contents and results are as follows:1.This paper gives an overview of the development,types and typical applications of mobile robot,illustrates the effect of path planning technology,and summarizes the research status of path planning technology in China and abroad.Several common modeling methods are introduced.This paper uses improved grid method for 3D environment modeling.To ensure the integrity of path search,this paper uses a search mode combining planar and visual domains.2.In the aspect of ant colony algorithm,this paper adopts a combined method of local pheromone update and global pheromone update,designs the heuristic function,and introduces rollback mechanism to avoid deadlock phenomenon.This paper conducts multiple simulation experiments separately at same start and end points in different terrains,and different start and end points in same terrain.Results show ant colony algorithm is suitable for 3D path planning.At the same time,the introduction of safety value calculation in environmental modeling stage can reduce average running time to 1/12 of calculating safety value in path planning stage.3.In the aspect of simulated annealing genetic algorithm,this paper introduces its notions and steps,and components such as encoding processing,initial population creation,fitness function and genetic operators.This paper does the following work:brings simulated annealing procedures respectively into the early and late running processes of algorithm;redesigns the roulette selection mechanism,and adopts best individual preservation strategy;under the condition that one chromosome has single variant gene,obtains the probability of one gene mutation,and limits the variation degree of gene.According to multiple simulation experiments conducted at same start and end points in different terrains,and different start and end points in same terrain.Through comparison it can be found that,in terms of avoiding obstacles,the optimal path outputted by simulated annealing genetic algorithm is better than the optimal path outputted by ant colony algorithm;the simulated annealing genetic algorithm has shorter average optimal path length than ant colony algorithm in multiple runs,the average path length is reduced by 6.85%.In terms of the operating time,the ant colony algorithm has slightly shorter operating time than simulated annealing genetic algorithm,the average operating time difference is in the range of 3 seconds.