RI Small: On Robot Motor Capability for Skill Learning

RI Small：论机器人运动能力的技能学习

• 批准号: 0916807
• 负责人: Chun-Sing Lee
• 金额: $ 45万
• 依托单位: Purdue University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2009
• 资助国家: 美国
• 起止时间: 2009-08-01 至 2017-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0916807&HistoricalAwards=false
• 关键词: RI; Small; Robot; Motor; Capability

Endowing humanoid robots with the ability of skill learning will enable them to be versatile and skillful in performing various tasks. The problem of transferring human skills to humanoid robots raises tremendous research interest in studying human and robot motor skills. Our research aims at developing a quantitative measure of robot motor capability of a humanoid root motor system for the application of transferring human skills to a humanoid robot. An in-depth study of basic intrinsic properties of robot motor capability based on information theory will be performed to derive a pseudo index of motor performance. This pseudo index of motor performance is derived from robot kinematics, dynamics, and control with the speed-accuracy constraint taken into consideration. With the speed-accuracy constraint, the motor performance of a robot is optimized to accomplish a task. The research results demonstrate an increased understanding of robot motor capability that shows the capability and limitation of a robot for learning skills from human demonstration. The project plans to verify the result on an existing humanoid robot. The project results will provide new insights in humanoid robot motor learning and control, and launch a new research direction on human-robot interaction when both humans and robots better know their respective motor capabilities. The broader impacts include incorporation of research results into existing graduate and undergraduate robotics courses, and outreach activities of a weeklong robotics summer camp and a lab open house for high-school students. The research results will be disseminated in professional conferences, workshops and archival journal publications.

赋予类人机器人技能学习能力，将使其能够灵活、熟练地完成各种任务。将人类技能转化为类人机器人的问题引起了人们对人类和机器人运动技能研究的极大兴趣。我们的研究旨在开发一种定量测量机器人运动能力的人形根电机系统，用于将人类技能转移到人形机器人上。基于信息论，深入研究机器人电机性能的基本内在特性，推导出机器人电机性能的伪指标。该伪电机性能指标是从机器人运动学、动力学和控制中得到的，同时考虑了速度-精度约束。在速度-精度约束下，优化机器人的电机性能以完成任务。研究结果表明了对机器人运动能力的进一步了解，表明了机器人从人类演示中学习技能的能力和局限性。该项目计划在现有的人形机器人上验证这一结果。项目成果将为仿人机器人的运动学习和控制提供新的见解，并在人类和机器人更好地了解各自的运动能力时，为人机交互开辟新的研究方向。更广泛的影响包括将研究成果纳入现有的研究生和本科生机器人课程，以及为高中生举办为期一周的机器人夏令营和实验室开放日活动。研究结果将在专业会议、讲习班和档案期刊出版物中传播。