Let Us Help You: The Development of Intelligent Robots as Effective Assistants for Society

让我们来帮助您：发展智能机器人作为社会的得力助手

• 批准号: RGPIN-2018-05811
• 负责人: Nejat, Goldie
• 金额: $ 4.01万
• 依托单位: University of Toronto
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2022
• 资助国家: 加拿大
• 起止时间: 2022-01-01 至 2023-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=759679
• 关键词: Let; Us; Help; Development; Intelligent

The world is currently experiencing a silver tsunami: rapid population aging. Robots will have a significant impact on society in order to meet the urgent needs of an aging population. In this research program, we will address the open challenge of incorporating intelligent service robots into society as long-term assistants to manage and perform multiple assistive activities in human-centred environments. This research will aim to make immediate breakthroughs to overcome the single task restriction of current service robots by equipping them with the perception and cognition needed to perform complex interactive behaviours. These behaviours will be pertinent to assisting multiple non-expert human users in numerous tasks in varying real-world settings. Aim 1 will develop embodied social intelligence for robots initiating and sustaining assistive interactions with humans. Sub-Aim 1.1 will uniquely identify and classify both human intent and affective (emotions and moods) cues to allow a robot to recognize complex human social behaviours during human-robot interaction (HRI). Both feature extraction and decision level fusion will be considered to classify social cues in order to determine accurate real-time multi-modal classification methods using probabilistic models. Sub-Aim 1.2 will address the challenge of a robot autonomously learning appropriate behaviours for assistance with multiple activities while also personalizing interactions. Learning will rely on the robot being able to perceive complex human social cues by considering the inherent uncertainty that exists when interacting with people in real-world settings.Aim 2 will develop novel HRI control architectures to allow multiple robots to cooperate with each other and users in order to provide multi-task assistance in social settings. Sub-Aim 2.1 will uniquely integrate social norms and person-centred care principles to facilitate effective learning of robot team cooperative assistive behaviours while ensuring focus is on user wellbeing. Sub-Aim 2.2 will develop a novel hierarchical multi-activity planning method for robot teams to allow real-time responsiveness to user, environment and activity changes in HRI settings. Activity planning will consider situations where robots will each have different goals or cooperate together to achieve the same common goal. The proposed Aims will be addressed within a centralized framework. New transformative knowledge will be generated in robot intelligence and control, multi-robot cooperation and HRI. The research will contribute to accelerating innovation within the robotics ecosystem, supporting the urgent need to strengthen made-in-Canada robotic technologies, and ensure Canada's role in this rapidly growing global market. A central outcome of this program will also be preparing HQP for successful careers in the multi-billion dollar service robotics industry.

世界目前正在经历一场银海啸：人口迅速老龄化。机器人将对社会产生重大影响，以满足人口老龄化的迫切需求。在这项研究计划中，我们将解决将智能服务机器人作为长期助手融入社会的开放性挑战，以管理和执行以人为中心的环境中的多种辅助活动。这项研究的目标是通过为当前服务机器人提供执行复杂交互行为所需的感知和认知来克服其单一任务的限制。这些行为将与帮助多个非专家人类用户在不同的现实世界环境中完成众多任务有关。目标1将为机器人开发具体的社会智能，启动和维持与人类的辅助互动。子目标1.1将唯一地识别和分类人类意图和情感（情感和情绪）线索，以允许机器人在人机交互（HRI）期间识别复杂的人类社会行为。特征提取和决策级融合将被认为是社会线索分类，以确定准确的实时多模态分类方法，使用概率模型。子目标1.2将解决机器人自主学习适当行为的挑战，以协助多种活动，同时还个性化交互。学习将依赖于机器人能够感知复杂的人类社会线索，通过考虑在现实世界中与人互动时存在的固有不确定性。Aim 2将开发新的HRI控制架构，允许多个机器人相互合作和用户合作，以便在社会环境中提供多任务协助。子目标2.1将独特地整合社会规范和以人为本的护理原则，以促进机器人团队合作辅助行为的有效学习，同时确保关注用户的福祉。子目标2.2将为机器人团队开发一种新的分层多活动规划方法，以实时响应HRI设置中的用户，环境和活动变化。活动规划将考虑机器人各自具有不同目标或共同合作以实现相同共同目标的情况。拟议的目标将在一个集中的框架内处理。在机器人智能与控制、多机器人协作和HRI等领域将产生新的变革性知识。该研究将有助于加速机器人生态系统内的创新，支持加强加拿大制造机器人技术的迫切需要，并确保加拿大在这个快速增长的全球市场中的作用。该计划的核心成果也将为HQP在数十亿美元的服务机器人行业的成功职业生涯做好准备。