プランクトン行動の新しいモデリング手法

概要

論文の詳細を見る
Many observatinal studies have shown that plankton are "intelligent" organisms, but in conventional simulation they have been treated as particles without any intentional movements. we have designed adaptive behavior models to simulate the movements of plankton, adopting the idea of Behavior-Based AI and the bottom-up approach. The models, which contain decision making networks as their adaptivity mechanisms, are based on observational studies of zooplankton and phytoplankton behaviors, and the networks decide the next movement or behavior according to physical and physiological conditions of plankton, which can be changed by the plankton's movement or behavior in turn. Although the networks are simple, the simulation results have shown that the behavior patterns emerged from the models are more similar to those of live plankton than the results from conventional stochastic models, and have suggested the importance of the models which can deal with animals' behaviors as emergent processes.
日本シミュレーション学会の論文
1998-03-15