Research On The Behavior Of Bee Swarm Based On Swarm Intelligence And Artificial Life

Swarm intelligence and artificial life have been developed rapidly in the last decade. As two artificial intelligence patterns, they open out the living phenomena and evolution rules in different hierarchies by simulating natural living phenomena, and provide a new kind of thought for complicated behaviors modeling and simulation. Research, development and practical application on artificial life and swarm intelligence will promote the exchange and development of life science, information science and systems science disciplines deeper, but also will produce a far-reaching impact on social science. Both research will expanded in wider areas and levels.In the paper, swarm intelligence and artificial life was selected as research methods, and the natural bee was chosen as the object of modeling virtual creature. Then try to establish a artificial creature system that have the characteristic owned by natural bees through the method of computer simulation, and show it by the graphical way. Firstly, the author studied a group of intelligence and artificial life theory and compared the common bionic algorithm. Secondly, through the natural behavior of the bees abstracted and summarized and analyzed several parts in detail about searching for nectar, recruiting bees for nectar, preparing to abandon several food of honey bees and the impact of seasonal variations, and then the author proposes a new algorithm based on inertia weight and dynamic vision. Experimental results show that the algorithm is effective and feasible. Inertia weight can make a bee keeping inertia and eliminate blindness bees choose the right direction. Dynamic vision can improve the convergence speed and accuracy of optimization algorithm.The author established a three-dimensional simulation platform for simulation of the behavior of bees in paper. The platform can set different environment variables. Changes in the behavior of bees can be presented on the simulation platform by changing the season and setting up obstacles. Finally, through the analysis of simulation results, further confirmed the algorithm based on inertia weight and dynamic vision can work well and the simulation effect is better.