Reduced Complexity Manipulation with the Parallel-Jaw Gripper: 2-Year Plan for Extended Research

为了使制造业保持竞争力，在设计产品时，必须考虑到诸如易组装、易掌握和易进给等因素。这项资助提供了一个更广泛的机会，以攻克令人兴奋的新研究课题，这些课题属于与最初的资助相同的一般研究领域：将传感器和执行器结合在一起，几乎没有自由度来抓取和定位部件。与平口夹爪相关的其他悬而未决的问题包括一种快速算法，用于估计给定多面体零件的两点摩擦抓取，并找到允许零件在主动枢轴伺服下旋转到所需最终姿势的抓取。我们还将研究零件在传送带上移动时如何定位和分类，以防止被动的栅栏顺序，以及一种在金属板上系统地输送振动受控的零件的方法。索尼在他们的APOS系统中使用了这个想法的变体，但他们没有编程的科学基础。这项新研究的目的就是建立这样一个科学基础。

For manufacturing industries to remain competitive, considerations such as the ease of assembly, graspability, and ease of feeding must be taken into account when products are designed. This grant provides an extended opportunity to attack exciting new research topics that fall within the same general research area as the original grant: combining sensors and actuators with few degrees of freedom to grasp and orient parts. Remaining open problems related to the parallel-jaw gripper include a fast algorithm for estimating 2 point frictional grasps of a given polyhedral part and finding grasps that will permit the part to be pivoted into the desired final pose under an active pivot servo. We will also investigate how parts can be oriented and sorted as they move on a conveyor belt against a passive sequence of fences and a method for systematically feeding parts with controlled vibrations on a metal plate. Sony uses a variant on this idea in their APOS system but they do not have a science base for programming it. The aim of this new research is to develop such a science base.