Coral Reef Algorithm And Its Application

Coral reef algorithm is a new biological meta heuristic algorithm, which is first proposed by S. Salcedo-Sanz et al in 2013, for optimizing multimodal function. Coral reef algorithm(CRO) is proposed based on the progress that form coral reefs and breed corals. As the algorithm is simple,easy and the search path is good,it is used to improve the mobile network and to solve the hard wind farm design question.Therefore,it has became a hot research topic in the field of Intelligent Heuristic algorithm, and attract a lot attention of scholars. However, the algorithm also has mang drawbacks such as early convergence speed, easy to fall into local optimum, the population diversity is not enough and so on. All of this limit the application of coral reef algorithm. Therefore, the coral reef algorithm need to be improved both in theory and application.This paper analyzes the shortcomings of the coral reef algorithm, and improved the algorithm from the update strategy and other aspects, and the improved algorithm is applied to the actual optimization problems. The main work of this paper includes the following three aspects:(1)Bring two populations into coral reef algorithm, one population update by coral reef algorithm, another population by the differential evolution algorithm.The two algorithms through information sharing mechanism to evolve populations in different directions.This strategy can increase the diversity of population, and enhance the global searching ability of the algorithm, avoid trapping into local optima by convergence to fast.(2)Put niche environment biology into the coral reef algorithm, proposed a coral reef algorithm based on niche environment. The basic idea of niche technology is to use the niche concept applied in evolutionary computation, divide the generation of evolutionary computation into several categories.Then select several large fitness individuals as an excellent representative and form a group. and then crossover and mutation to produce a new generation of individuals, so that we can effectively increase the diversity of population, enhance the global search capability of the algorithm, and the improved algorithm is applied to solve the 0-1 knapsack problem.(3)The paper put the improved algorithm into solving robot path planning in static environment, in order to improve the application range of the coral reef algorithm.