arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2004.10558cs.NEcs.MA
收藏

一个协同物理任务的演化二元策略

Evolving Dyadic Strategies for a Cooperative Physical Task

Saber Sheybani, Eduardo J. Izquierdo, Eatai Roth

摘要:

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。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑