Department of Intermedia, Art and Science, School of Fundamental Science and Technology, Waseda University, Ohkubo 3-4-1, Shinjuku-ku, Tokyo, 169-8555, Japan.
Biosystems. 2022 Dec;222:104776. doi: 10.1016/j.biosystems.2022.104776. Epub 2022 Sep 19.
Metaheuristic algorithms are a method of searching for a suboptimal solution and are implemented by a large number of agents, such as a swarm of animals. In this method, it is very important to adjust the relationships and/or interplay between individual freedoms and constraints as a group. The phenomena related to this interplay are strongly relevant for self-organized criticality showing a power law distribution. Since individuals' independent behavior and translation as a group coexist in a swarm of soldier crabs, we conducted an experiment involving soldier crabs to estimate whether a power law distribution results from that interplay. The engine of translation was implemented by the directed movement of robot crabs. Soldier crabs confronting moving robot crabs were observed in the experiment. Here, we show that the interaction between robots and soldier crabs entails a Lévy walk characterized by a power law distribution of step length as well as the real swarm. This result suggests that soldier crabs could be attracted by the directed motion of robots and that combining a random walk with ballistic movement can realize a Lévy walk.
元启发式算法是一种搜索次优解的方法，通过大量的代理（如动物群）来实现。在这种方法中，调整个体自由和群体约束之间的关系和/或相互作用非常重要。与这种相互作用相关的现象与自组织临界性密切相关，表现出幂律分布。由于兵蟹群中个体的独立行为和群体翻译同时存在，我们进行了一项涉及兵蟹的实验，以评估这种相互作用是否会导致幂律分布。翻译的引擎是通过机器人蟹的定向运动来实现的。在实验中观察到兵蟹面对移动的机器人蟹的情况。在这里，我们表明机器人蟹和兵蟹之间的相互作用需要一种 Lévy 游走，其特征是步长的幂律分布以及真实的群体。这一结果表明，兵蟹可能会被机器人蟹的定向运动所吸引，并且随机游走和弹道运动的结合可以实现 Lévy 游走。
