Font Size: a A A

AGV Path Planning And Application Based On Swarm Intelligence Algorithm

Posted on:2022-05-07Degree:MasterType:Thesis
Country:ChinaCandidate:H ChenFull Text:PDF
GTID:2518306530455584Subject:Master of Engineering
Abstract/Summary:PDF Full Text Request
AGV path planning in our country is in the stage of rapid development,which has been widely used in autonomous mobile robots.Among them,intelligent algorithm is the core of AGV path planning,but there are still shortcomings in the existing swarm intelligence algorithm.The main work contents in optimizing algorithm are as follows:Firstly,this paper selects a fixed scene as the experimental object,compares and analyzes several existing map building methods,and finds out their advantages and disadvantages.The actual environment is simplified as a map with only feature information points,and then the map path planning problem with only information points is transformed into a traveling salesman problem of 52 cities and 127 cities through the idea of equivalence.By studying the simulation situation after modeling,the path planning algorithm of robot is selected.Secondly,this article compares the experimental data of GA algorithm,ACO algorithm,PSO algorithm and FWA algorithm in dealing with the same traveling salesman problem.And then finds that the performance of each algorithm has defects,so the relatively better FWA algorithm is selected to improve.The main problem with FWA algorithm is that there is no information interaction among individuals,which leads to the problem of obtaining the local optimal solution prematurely.Therefore,information interaction between individuals is introduced to enhance its explosive effect to obtain the EFWA algorithm.After the improvement,the mean error of EFWA algorithm is as low as 0.24% in the 52#TSP problem and 1.45% in the 127#TSP problem,both of which are better than all the values before the improvement.The data shows that EFWA algorithm is better.Finally,the experimental data and results of this article are summarized.Through further analysis of the research results,it is found that the algorithm can be applied to more situations,and similar traveling salesman problems can adopt the EFWA algorithm to obtain the shortest path.The personal safety of the staff is always threatened by the high voltage working environment of the substation.The machine inspection can avoid the occurrence of accidents commendably,and because of its relatively fixed environment,it can be well applied to the improved EFWA algorithm in this paper,which has a good effect in the actual testing.In order to facilitate easier manual operation,this paper designs an app that can control the robot to complete the task in real time and test it based on the theoretical support.
Keywords/Search Tags:swarm intelligence algorithm, AGV path planning, environment modeling, transformer substation
PDF Full Text Request
Related items