Research On Path Planning Of Mobile Robot Based On Improved Ant Colony Optimization

In recent years,with the development and progress of artificial intelligence technology,it has become more and more important in social life and production,thus attracting the favor of the majority of scholars,and has become a research hotspot.There are many research branches,among which,the research of intelligent mobile robot is also a hot issue at the present stage.Intelligent mobile robots are widely used in important fields such as warehousing and logistics,unmanned driving,navigation system and so on,involving national defense,military,people’s livelihood,economy and so on.Therefore,improving the work efficiency of mobile robots is an important and necessary content.Scholars at home and abroad have done a lot of research on robot path planning algorithms,common algorithms include A* algorithm,particle swarm optimization algorithm,ant colony optimization,and so on.Among them,ant colony optimization has the characteristics of robustness and parallelism,which is favored by a large number of scholars.With the deepening of research,it is found that ant colony optimization also has the disadvantages of slow convergence speed,low search efficiency,easy to fall into local optimal and unable to avoid dynamic obstacles.Based on this,the path planning of mobile robots using improved ant colony optimization is studied.The main research contents of this paper are as follows:1.In this paper,the principle and procedure of traditional ant colony optimization are introduced,and the advantages and disadvantages of traditional ant colony optimization in path planning of mobile robot are discussed.Through the analysis of the traditional ant colony optimization,it is concluded that the ant colony optimization simulates the foraging process of ants in nature and transmits information through pheromones,which is a positive feedback algorithm,so it has the advantages of strong robustness,but it also has the disadvantages of long time,low efficiency,and easy to fall into the local optimum.2.Aiming at the problems of slow convergence speed,low search efficiency and easy to fall into local optimal of traditional ant colony optimization.In the third chapter,an improved ant colony optimization is proposed,which includes: The grid environment was divided into two parts with the starting point and the ending point line as the boundary,and the initial pheromone distribution of the normal distribution was used,in which the initial pheromone distribution of all nodes was correlated with their positions,so as to ensure that different nodes were allocated different initial pheromones according to their Euler distance from the starting point to the ending point line,and the advantage of the initial pheromone differential distribution was fully reflected,and the initial blindness of the algorithm was reduced greatly.Based on the relationship between average path and shortest path of each generation,the whole search process is divided into two stages: the early stage and the late stage.The higher value of volatile factor in the early stage is convenient for ant colony to carry out global search and reduce the possibility of local optimal.In the late stage,the smaller volatile factor is selected to accelerate the algorithm convergence.The shortest path is obtained by further optimizing the redundant path,which makes the algorithm faster and more stable.The feasibility and authenticity of the improved ant colony optimization are verified by the final simulation results.3.In order to further improve the efficiency of the improved ant colony optimization,a fusion algorithm is proposed based on the improved ant colony optimization in Chapter 3.The improved ant colony optimization is integrated with genetic algorithm,and the advantages of ant colony optimization and genetic algorithm are combined to improve the adaptability of fusion algorithm to the environment.In view of the shortcomings of single ant colony optimization,such as many redundant nodes,slow convergence speed and low efficiency,the optimal strategy and genetic region strategy were proposed.The high-quality parent path in each generation path was selected,and the search range of offspring was delimited by the parent path,which narrowed the search range of offspring ants and reduced the occurrence of redundant nodes,thus improving the convergence speed and efficiency of the algorithm.The feasibility and effectiveness of the ant colony genetic fusion algorithm were verified by simulation experiments.4.Due to the limitations of the traditional ant colony optimization,it has low adaptability to the dynamic environment.Therefore,when the obstacles in the environment change,it cannot avoid the obstacles well,and the timeliness is poor.Aiming at the dynamic path planning problem of mobile robots,a "wait-falling-local planning" strategy is proposed on the basis of the improved ant colony optimization proposed in this paper: Firstly,the improved ant colony optimization is used to plan a global path.The robot advances along the shortest path searched.If it detects a collision with a dynamic obstacle,the next action of the robot is adjusted by judging the motion state of the obstacle,namely local path planning.This strategy can effectively avoid the collision between robot and dynamic obstacles and improve the efficiency of ant colony optimization in mobile robot dynamic path planning.5.Finally,the experiment platform of mobile robot path planning was built to verify the obstacle avoidance strategies of the improved ant colony optimization in static path and dynamic path planning for mobile robot.