Integration of high-level cognitive functions in an autonomous robot control architecture

在自主机器人控制架构中集成高级认知功能

• 批准号: 203577-2011
• 负责人: Michaud, François
• 金额: $ 3.64万
• 依托单位: Université de Sherbrooke
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2014
• 资助国家: 加拿大
• 起止时间: 2014-01-01 至 2015-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=546365
• 关键词: Integration; level; cognitive; functions; autonomous

Over the last decade, mobile robots have evolved from being rigid, two-dimensional navigation platforms, to compliant and interactive machines that can accomplish more complex and sophisticated tasks, assuming that they can exploit their advanced capabilities intelligently. The ultimate goal is for such robot to be able to decide on its own what, when and how they can fulfill their duties in a secure, efficient and productive manner, with minimal human intervention. I have been developing a mobile platform that brings the necessary compliance, robustness and adaptiveness to handle autonomously the unconstrained and unpredictable situations of the real world. The platform is a compliant omnidirectional humanoid mobile robot capable of natural reciprocal interaction (motion, navigation, auditory, visual, language, touch, affect) with humans, and provides an excep- tional and advanced testbed to address the integration and coordination issues that a robot control architecture must face to operate in real life settings. The focus of this proposal is on the memorization and categorization of the perception and the situations experienced by the robot, 2 key high-level cognitve functions. Memory is fundamental to reasoning, learning and cognition, but it still remains a poorly understood but required mecha- nism for robots having to understand its interactions in dynamic environments. I will study and develop approa- ches that will be integrated into a hybrid behavior-based architecture, to: exploit the concepts of short-term and long-term memory to achieve real-time processing and to derive a spatio-temporal representation of the robot's experiences from internal and perceived data; memorize and categorize events through the combination of visual, audible and tactile information; regulate satisfaction of the robot's intentions using emotional signaling, which can in term be used to categorize behavioral strategies. These approaches will be evaluated in the context of making an interactive robot assistant that will have to operate in our facility for the longest time possible. Performance will be measured in terms of the added capabilities brought by the high-level control mechanisms, and from the robot's overall ability to operate autonomously in open environments.

在过去的十年里，移动机器人已经从僵化的二维导航平台演变为合规和互动的机器，可以完成更复杂和复杂的任务，前提是它们可以智能地利用自己的高级能力。最终目标是这样的机器人能够自己决定什么、何时以及如何以安全、高效和多产的方式履行职责，而只需最少的人工干预。我一直在开发一个移动平台，它带来了必要的合规性、健壮性和适应性，可以自主处理现实世界中不受限制和不可预测的情况。该平台是一个顺应的全方位仿人移动机器人，能够与人类进行自然的交互(运动、导航、听觉、视觉、语言、触摸、情感)，并提供一个额外的高级试验台来解决机器人控制体系结构在现实生活环境中操作必须面对的集成和协调问题。该方案的重点是对机器人所经历的感知和情景的记忆和分类，这是两个关键的高级认知功能。记忆是推理、学习和认知的基础，但它仍然是一个鲜为人知的机制，但对于机器人来说，它必须理解它在动态环境中的相互作用。我将研究和开发将被整合到基于混合行为的架构中的方法，以：利用短期和长期记忆的概念来实现实时处理，并从内部和感知数据中推导出机器人经验的时空表示；通过视觉、听觉和触觉信息的组合来记忆事件并对事件进行分类；使用情感信号来调节机器人意图的满足，这在术语上可以用于对行为策略进行分类。这些方法将在制造交互式机器人助手的背景下进行评估，该助手必须在我们的设施中运行尽可能长的时间。性能将通过高级控制机制带来的附加能力以及机器人在开放环境中自主操作的整体能力来衡量。