Coordination of a Human Arm and a Robot Manipulator

这是一个为期三年的持续资助项目的第一年。研究了机械臂与人类操作者手臂协同工作的协调机制。通过机器人机械臂与操作者手臂的协调，可以通过人类操作者的智能接近非结构化环境，而物体的负载可以由机械手和人类操作者共同承担。采用三步法研究了与柔顺运动、反射运动和运动学习相关的机械臂协调问题。将柔性运动方法应用于机器人与人类操作者的交互中，以使机器人按照人类操作者的运动进行运动。自反运动是本研究中有待发展的一种新机制。其目的是当操作者通过被协作处理的物体对机械手施加提示力时，使机械手对人类操作者的意图作出适当的响应。其结果是，机械臂将主动移动，而人类操作员将不需要不施加连续的提示力给机械臂。另一个正在探索的新机制是动作学习。这使得机械臂能够在柔顺和反射运动阶段学习与人类操作者最佳协调的轨迹。因此，当重复相同的操作任务时，机械手可以使用学习到的轨迹，进一步减少了人体手臂的负荷。该研究将为人类手臂-机械臂协调技术的发展提供理论基础，从而极大地扩展机器人的应用范围，有利于制造业、服务业和建筑业。

Zheng This is the first year funding of a three-year continuing award. The research studies coordination mechanisms for a manipulator to work cooperatively with a human operator's arm. By coordinating a robot manipulator with the operator's arm, unstructured environments can be approached through the intelligence of the human operator, while the load of the object can be shared by the manipulator and the human operator. A three-step approach is taken to study the arm-manipulator coordination problem relative to compliant motion, reflexive motion, and motion learning. A compliant motion approach is applied to the interaction of a robot with a human operator in order to make the robot move in accordance with motion of the human operator. Reflexive motion is a new mechanism to be developed in this research. The purpose is for the manipulator to respond appropriately to the intention of the human operator when the operator applies a cueing force to the manipulator via the object being cooperatively handled. The result is that the manipulator will move actively and the human operator will not need not to apply continuous cueing forces to the manipulator. Another new mechanism being explored is motion learning. This allows a manipulator to learn a trajectory that best coordinates with the human operator during the compliant and reflexive motion stages. As a result, when the same manipulation task is repeated, the manipulator can use the learned trajectory, further reducing the load on the human arm. This research will lead to a theoretical foundation for the development of human arm- manipulator coordination technologies which could greatly extend the range of robotic applications with benefit to manufacturing, service, and construction industries.