Ant Colony Algorithm And Its Application In The Continuous Space Optimization Problems

The ant colony algorithm is a kind of stochastic explorative algorithms. As the same of other kind of stimulated evolutionary algorithms, it finds the best solution of optimization problem by making uses of the evolutionary procedure of a set of cooperating agents of candidate solutions. It has showed a great deal of salient character and performed great value in its application. Choosing the analysis of charater of Ant System(the basic algorithms of ant colony) as the research background, this paper focuses on the model, the behavior, the characteristics and its improvements, and raises a model of ant colony algorithm on general function optimization problems. Finally we discuss the salient character of ant colony algorithm on the continuous space(general function) optimization problems, and propose the region of further investigation. The main contents are composed of the following parts:1. The biological mechanism, the development and the character of the ant colony algorithm are outline. The current situations of reseach and application of the ant colony algorithm is introduced. The superiority in solving combinatorial optimization problems and the shortages in solving the continuous space optimization problems are presented.2. The principle, the model, the characteristics and the management about the basic algorithms of ant colony(Ant System) are also presented. The reasionable selection about the parameters is discussed in emulated test. The characteristics about the ant colony algorithm is analysed thoroughly.3. For the shortages of the basic algorithms of ant colony, a list of improvements have been referenced and presented, so the characteristics of ant colony algorithm to be made more excellently.4. The model of ant colony algorithm for general function optimization problem is presented. A feasible way is presented for applying the algorithms to practice. The result about the emulatedtest is demonstrated that the characteristics of ant colony algorithm for general function optimization problem are salient, and it is wealthy to make further researches.5. To a system which is SISO and constant, a methed to get the definite scopes of weighting matrix Q and state feedback matrix K is presented. By making use of the ant colony algorithm in the continuous space optimization problems which is proposed in this paper, the discrete LQ inverse problem is solved. The results of the computer simulation show that the ant colony algorithm is a valid method for solving the control optimization problems.6. The ant colony algorithm is a kind of stochastic explorative algorithms which has showed many excellent characteristics. It is demonstrated that the ant colony algorithm is more adaptivable to the genetic algorithms and the simulated annealing algorithms which were fasionable for a time. Although some successful applications have been presented, it also has many problems for solving and making further investigate.