Vision-Based Control of Dextrous Manipulators in High Precision Tasks

EEC-9700702 摘要 该奖项提供资金圣母大学的指导下，博士迈克尔M。Stanisic，以支持一个名为“高精度任务中基于视觉的灵巧机械手控制”的联合研究课程开发项目。“这个项目将为圣母大学的高年级本科生和入门级研究生开发一个四门课程的序列。 该序列将教学生设计和控制一类新的机械手，能够制造任务，这是非常困难或昂贵的实现与现有技术。 该系列中的顶点课程涉及学生编程一个用于执行此类制造任务的操作系统，这些任务来自工业咨询小组，这是该项目的关键要素。该提案的研究方面扩展了所涉及的技术，开发了一种新型的无奇异性机械手，并将免校准，基于视觉的控制算法扩展到易于使用的“点和点击”任务编程手段。

EEC-9700702 Abstract The award provides funding to the University of Notre Dame under the direction of Dr. Michael M. Stanisic, for the support of a Combined Research-Curriculum Development project entitled, "Vision-Based Control of Dextrous Manipulators in High Precision Tasks." This project will develop a four-course sequence for upper-level undergraduate and entry-level graduate students at the University of Notre Dame. The sequence will teach students about the design and control of a new class of dextrous manipulators, capable of manufacturing tasks which are prohibitively difficult or costly to achieve with existing technology. The capstone course in the sequence involves students programming a dextrous manipulatory to perform such manufacturing tasks, tasks drawn from an Industrial Advisory Panel, which is a key element of this project. The research aspect of the proposal extends the technologies involved, developing a new type of singularity-free manipulator, as well as extending the calibration-free, vision-based control algorithm to an easily used "point-and-click" means of task programming.