Robots might not act according to human expectations if they cannot anticipate how people make sense of a situation and what behavior they consider appropriate in some given circumstances. In many cases, understanding, expectations and behavior are constrained, if not driven, by culture, and a robot that knows about human culture could improve the quality level of human–robot interaction. Can we share human culture with a robot? Can we provide robots with formal representations of different cultures? In this paper, we discuss the (elusive) notion of culture and propose an approach based on the notion of trait which, we argue, permits us to build formal modules suitable to represent culture (broadly understood) in a robot architecture. We distinguish the types of traits that such modules should contain, namely behavior, knowledge, rule and interpretation traits, and how they could be organized. We identify the interpretation process that maps situations to specific knowledge traits, called scenarios, as a key component of the trait-based culture module. Finally, we describe how culture modules can be integrated in an existing architecture, and discuss three use cases to exemplify the advantages of having a culture module in the robot architecture highlighting surprising potentialities.

中文翻译：

---

具有文化能力的机器人的基于特征的模块

如果机器人无法预测人们如何理解情况以及他们认为在某些特定情况下哪些行为是合适的，那么它们可能不会按照人类的期望行事。在许多情况下，理解、期望和行为受到文化的限制，如果不是驱动的话，了解人类文化的机器人可以提高人机交互的质量水平。我们可以与机器人分享人类文化吗？我们能否为机器人提供不同文化的正式表现形式？在本文中，我们讨论了（难以捉摸的）文化概念，并提出了一种基于特征概念的方法，我们认为，这种方法允许我们构建适合在机器人架构中表示文化（广义理解）的正式模块。我们区分这些模块应该包含的特征类型，即行为、知识、规则和解释特征，以及如何组织它们。我们确定了将情境映射到特定知识特征（称为场景）的解释过程，作为基于特征的文化模块的关键组成部分。最后，我们描述了如何将文化模块集成到现有架构中，并讨论了三个用例来举例说明在机器人架构中拥有文化模块的优势，突出了令人惊讶的潜力。