Evolving Dyadic Strategies for a Cooperative Physical Task
Evolving Dyadic Strategies for a Cooperative Physical Task
复制标题
合作物理任务的不断发展的二元策略
DOI：
--
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Eatai Roth
中科院分区：
文献类型：
--
作者：
Saber Sheybani;E. Izquierdo;Eatai Roth
许多合作的体力任务需要个人扮演专门的角色（例如，领导者-追随者）。人类是熟练的合作者，天生就能协调这些角色和角色之间的转换。然而，角色是如何被委派和重新分配的，人们还没有很好地理解。使用遗传算法，我们进化模拟代理，探索一个空间的可行的角色切换政策。应用这些切换策略在一个合作的手动任务，代理处理视觉和触觉提示，以决定何时切换角色。然后，我们分析进化的虚拟人口的属性通常与合作：负载分担和时间协调。我们发现，表现最好的二人组表现出较高的时间协调（反同步）。反过来，反同步与合作代理的参数之间的对称性相关。这些模拟提供了假设，人类合作者如何可能调解角色的二元任务。
Many cooperative physical tasks require that individuals play specialized roles (e.g., leader-follower). Humans are adept cooperators, negotiating these roles and transitions between roles innately. Yet how roles are delegated and reassigned is not well understood. Using a genetic algorithm, we evolve simulated agents to explore a space of feasible role-switching policies. Applying these switching policies in a cooperative manual task, agents process visual and haptic cues to decide when to switch roles. We then analyze the evolved virtual population for attributes typically associated with cooperation: load sharing and temporal coordination. We find that the best performing dyads exhibit high temporal coordination (anti-synchrony). And in turn, anti-synchrony is correlated to symmetry between the parameters of the cooperative agents. These simulations furnish hypotheses as to how human cooperators might mediate roles in dyadic tasks.