The Study, Based On The Animat Visual Model Of Intelligent Optimization Mode

Optimization problem exists in actual life widely, and the effect of optimization becomes more and more important along with the increasing need of economization on energy and environmental protection, etc. Optimization methods, the theory and algorithm of solving optimization problems, can be divided into two parts, the traditional optimization algorithms and intelligent optimization algorithms. The traditional optimization method, mainly based on derivative information, is also called as gradient optimization method. This kind of method mainly obtain the solution information by the derivation of objective function, it also need definite problem model, nice differentiability and continuity to help the problem solving. But in fact, according to the upward trend of the complexity, diversity and indeterminacy of actual optimization problem, the application area of traditional optimization method becomes more and more narrow. The intelligent optimization algorithm emerges as the times require, it overcomes those disadvantages and has characteristices as universality, adaptability, parallelity and so on, and to be used successfully in optimization area to solve the complex optimization problem in actual life. But there are also some disadvantages with the general intelligent optimization algorithm, such as lacking strict theoretic demonstration, having no uniform framework, only can make a concrete analysis of concrete problems, local minimum, lower precision, time consuming, and prematurity phenomenon. Therefor, it is an excellent topic of research and application.The essential of general intelligent opimization strategies that have been used is the simulation of biological phenomena or natrual law. The theory of animat is a kind of representation of artificial life, it has many characteristices of complex system such as autonomy, adaptability, emergence and so on. An optimization mode based on animat theory is presented. It simulates animal's path seeking ability in complex and unknown environment. An animat structure mainly consisted of the vision model and reacting behavior mechanism of animals is implemented via object-oriented analysis method, which provides us a flexible and stable optimization mode. The developed Powell method is used in the vision model of animat, and then the functions of the optimization mode is enlarged. Contraposed to multi-modal globle optimization problem, experiments on typical test functions are carried out to...