Capture Regions for Grasping, Manipulating and Re-orienting Parts

摘要IIS-9907009 Ponce，让·伊利诺伊大学香槟分校108,395-12美元。抓取、操纵和重新定位PARTS的CAPTURE区域是为期三年的连续奖项的第一年。PI解决了使用简单机器人设备的一系列接触来操纵对象的问题，这些设备可能具有离散或连续的自由度。规划整个物体轨迹的问题归结为构造一个原子机器人在系统的平衡状态之间运动的序列。该方法基于构造每个平衡点的捕捉区域，即物体构形空间的最大子集，使得在这个子集内开始的任何运动都保证以平衡位形结束。具体地说，在操作任务的执行期间，不假设保持接触；也不存在任何预测平衡之间的对象运动的尝试。这种方法将被应用于三个问题：(1)在存在摩擦效应和较大位置不确定性的情况下的抓取和手部操作；(2)移动机器人在避开障碍物的同时推动对象的运动规划；(3)通过将部件放置在让人联想到弹子机的简单机构上来重新定位部件。将通过使用实际的机器人设备进行实验来评估进展，这些设备包括四个游牧超级童子军II移动机器人以及由伊利诺伊大学开发的可重新配置的抓取器和拟议的弹子机的原型。提出的研究主要应用于制造业，包括零件进给和工业装配。更长期的应用包括控制家用机器人团队，让它们在房子周围做家务。

Abstract IIS-9907009Ponce, JeanUniversity of Illinois at Urbana-Champaign$108,395 - 12 mos.CAPTURE REGIONS FOR GRASPING, MANIPULATING AND RE-ORIENTING PARTSThis is the first year of a three-year continuing award. The PI addresses the problem of manipulating an object using a succession of contacts with simple robotic devices that may have discrete or continuous degrees of freedom. The problem of planning the overall object trajectory is reduced to constructing a sequence of atomic robot motions between equilibrium states of the system. The proposed approach is based on constructing the capture region of each equilibrium, i.e., the maximal subset of the object's configurationspace such that any motion starting within this subset is guaranteed to end up in the equilibrium configuration. In particular, contact is not assumed to maintain during the execution of the manipulation task; nor is there any attempt to predict the object motion between equilibria. This approach will be applied to three problems: (1) grasping and in-hand manipulation in the presence of frictional effects and large position uncertainties, (2) motion planning for mobile robots pushing an object while avoiding obstacles, and (3) re-orienting parts by dropping them on a simple mechanism reminiscent of Pachinko machines. Progress will be assessed by conducting experiments with actual robotic devices, including four Nomadic Super Scout II mobile robots and prototypes of the reconfigurable gripper and of the proposed Pachinko machine developed at the University of Illinois. The main applications of the proposed research are in manufacturing, including part feeding and industrial assembly. Longer-term applications include the control of teams of domestic robots as they perform chores around a house.