Font Size: a A A

The Improved Artificial Bee Colony Algorithm And Its Application In Calculating Seven Parameters Of Coordinate Transformation

Posted on:2017-02-21Degree:MasterType:Thesis
Country:ChinaCandidate:D DuFull Text:PDF
GTID:2308330485964138Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
Artificial bee colony algorithm is a relatively new emergence of swarm intelligence optimization search algorithm, its basic principle is to imitate swarm intelligence which is showed in the process of collecting nectar behavior of the bee groups to realize optimizing the actual problem. In 2005, based on years of research on the principles of honey bees, the Turkey academics Karaboga proposed the artificial bee colony algorithm. Because of the artificial bee colony algorithm have the advantage of is easy to understand, benefit from the implementation of robust, have little parameters, and so on. Soon attracted the attention of many researchers. But as a new kind of algorithm, the model of the algorithm is not very perfect, it’s applications in optimizing some complex practical problems is still in the preliminary stage. So, people are constantly exploring the algorithm’s application in the field of new, is no longer just limited to the function of the application of optimization. From the current status of research, the artificial bee colony algorithm has been successfully applied in robot path planning, network routing, scheduling, wireless sensor network node deployment, parameter optimization and image segmentation, and other fields, and new applications scene is constantly being proposed. However, in the process of applied to practical problems, artificial bee colony algorithm’s shortage of place is reflected out, the main problem is the speed of convergence is slow, the convergence precision is low, and easily to fall into local optimum, which makes it in the results of solving some of the problems also are not very satisfied by people. Therefore, this algorithm would be to improved research of many scholars.In this paper, based on basic artificial bee colony algorithm, aiming at the above-mentioned deficiencies of artificial bee colony algorithm, combined with practical problem solving coordinate transformation seven parameters, the paper proposed two improvements of basic artificial bee colony algorithm, then applying it to calculate seven parameters of coordinate conversion. The main contents are as follows:Firstly, the paper describes the basic theory and the key idea of the artificial bee colony algorithm, then say about the basic steps of the algorithm, introducing the basic theoretical knowledge, the introduction of the type of coordinate system as well as the method of transformation of coordinate. Then in connection with the initial solution of artificial bee colony algorithm is generated by random, has a certain blindness problems, also combined with the calculation problem of seven-parameter in coordinate transformation:three randomly selected coincident points computed seven parameters have a significant part of the coordinate transformation is not satisfied with the accuracy requirements, and then go on to improve the characteristics has little significance, so we can exclude the seven parameters which calculating coordinate conversion accuracy is not precise and deleting corresponding coincident points generated in initial solution, in the way will speed up the rate of algorithm’s convergence when it is applied to the calculation of seven-parameter in coordinate transformation. Thus for the subsequent calculations will get better and better.Then, artificial bee colony algorithm in the stage of the onlooker bees used the same search strategy with employed bees, ignoring the relationship between the individual, then the algorithm is easily to fall into local optimum shortcomings, also combined with the seven-parameter coordinate conversion calculation, The number of coincidence point and the external polygon area which is formed by coincidence point in coordinate system, can influence the precision of coordinate transformation which is calculated by seven-parameter, this paper proposes in the stage of observed bees adding the evaluate of the number of coincidence point and the external polygon area which is formed by coincidence point in coordinate system, This allows the algorithm can find the globally optimal solution better.Finally, to prove the priority precision of improved artificial bee colony algorithm when is used in the calculation of seven parameters in coordinate transformation. This paper carried out the comparative experiments of the improved artificial bee colony algorithm and the based artificial bee colony algorithm when applied to the calculate of seven parameters in coordinate system transformation, and contrast experiment between the improved artificial bee colony algorithm and the present stage common method of the calculation of seven parameters in coordinate system transformation, the experimental results show that the seven parameters calculated by the proposed artificial bee colony algorithm which is carried out by this paper can find out highly coordinate transformation precision and high stability.
Keywords/Search Tags:Artificial bee colony algorithm, Initialization of solution, Onlooker bee phase, Seven parameters method, Coordinate trartsformation
PDF Full Text Request
Related items