CRII: CHS: Human-Robot Collaboration in Special Education: A Robot that Learns Service Delivery from Teachers' Demonstrations

这个项目的重点是通过使用机器人技术来改善有特殊需要的儿童的生活质量。2011年，美国13%的入学儿童需要接受某种形式的特殊教育；2010年，美国每68名儿童中就有1名被诊断患有某种形式的自闭症谱系障碍（ASD），这是一种几乎总是需要特殊教育的发育状况。在过去的十年中，大量的机器人研究表明，许多自闭症儿童对机器人玩具有浓厚的兴趣，这表明机器人是这些儿童特殊教育的潜在有价值的工具。但是，有权批准在特殊教育中使用这种设备的临床社区和教育工作者仍然不相信。造成这一差距的一个原因是，这一领域的机器人研究并没有很强的关注有效性。PI最近完成了一项涉及自闭症儿童的初步研究，证实了通过人机交互（HRI）提供特殊教育服务的临床可行性。她目前项目的目标是建立一个研究项目，以这些发现为基础，开发一个框架，包括方法和算法，为自闭症儿童和其他类似性质的疾病提供机器人介导的特殊教育服务。PI的愿景是为教育工作者与机器人合作提供服务的未来奠定基础，从而通过提高特殊教育服务的有效性产生重大影响，同时减轻教育工作者在服务提供过程中的负担。本研究提出了一种“从示范中学习”（LfD）的方法，即机器人从人类教育工作者在真实的特殊教育场景中进行的一系列示范中学习服务交付。当人类教育工作者对机器人的能力有足够的信心时，他/她将启动一个协作服务交付场景，在这个场景中，机器人自主地提供特定教育服务的一些步骤，但在它对自己的感知和/或计划行动缺乏信心的情况下，它会请求教育工作者的帮助。多模态活动识别框架将提供有关特殊需要儿童在与机器人接触时的反应和活动的信息。为了促进教育工作者和机器人的互动，将组织一个由教育工作者和临床医生组成的焦点小组，以满足他们在设计特殊教育服务交付机器人的控制界面方面的需求和期望。一项人力资源研究所的研究将在现实环境中测试所提议的框架的有效性。能够与人类教育者合作提供特殊教育服务的机器人的设计和实现将为推进机器人学习（特别是高级概念学习）、机器人视觉和活动识别以及HRI方面的最新技术提供许多机会。

This project focuses on improving the quality of life for children with special needs through the use of robot technologies. In 2011, 13% of children enrolled in schools in the United States required some form of special education; in 2010, 1 in 68 children in the U.S. was diagnosed with some form of autism spectrum disorder (ASD), a developmental condition that almost always triggers the need for special education. A significant amount of robotics research over the past decade has indicated that many children with ASD have a strong interest in robotic toys, and suggested that robots are potentially valuable tools in special education for these children. But the clinical community and educators, who have the authority to approve the use of such devices in special education, remain unconvinced. One reason for this gap is that the robotics research in this domain has not had a strong focus on effectiveness. The PI recently completed a preliminary study involving children with ASD that confirmed the clinical feasibility of special education service delivery through human-robot interaction (HRI). Her goal in the current project is to establish a research program that builds on these findings to develop a framework, including methods and algorithms, for robot-mediated special education service delivery to children with ASD and other disorders of a similar nature. The PI's vision is to lay the foundations for a future where educators deliver services in collaboration with robots, thereby making a significant impact by increasing the effectiveness of special education services while reducing the burden on the educator during the service delivery process.This research proposes a "Learning from Demonstration" (LfD) approach where the robot learns service delivery from a series of demonstrations by human educators in real special education scenarios. When the human educator has sufficient confidence in the robot's ability, s/he will initiate a collaborative service delivery scenario in which the robot autonomously delivers some steps of a specific educational service but asks for the educator's assistances in cases where it has low confidence in its perception and/or planned actions. A multi-modal activity recognition framework will provide information about the responses and activities of children with special needs as they engage with the robot. To facilitate educator/robot interaction, a focus group will be organized with educators and clinicians to accommodate their needs and expectations in the design of the control interface for the special education service delivery robot. An HRI study will test the effectiveness of the proposed framework in real-world settings. The design and implementation of robots that can deliver special education services in collaboration with a human educator will open up many opportunities to advance the state of the art in robot learning (especially high-level concept learning), robot vision and activity recognition, and HRI.