Toward automatic robot instruction from perception-recognizing a grasp from observation

被引：38

作者：

Kang, Sing Bing ^{[1
]}

Ikeuchi, Katsushi ^{[1
]}

机构：

[1] Carnegie Mellon Univ, Pittsburgh, United States

来源：

IEEE Transactions on Robotics and Automation | 1993年 / 9卷 / 04期

基金：

美国国家科学基金会;

关键词：

Automation - Manipulators - Robot programming - Robotic assembly;

D O I：

10.1109/70.246054

中图分类号：

学科分类号：

摘要：

Many of the tasks that are potential candidates for automation involve grasping. The authors are interested in the programming of robots to perform grasping tasks. To do this, the assembly plan from observation (APO) paradigm is adopted, where the key idea is to enable a system to observe a human performing a grasping task, understand it, and perform the task with minimal human intervention. A grasping task is composed of three phases; pregrasp phase, static grasp phase, and manipulation phase. The first step in recognizing a grasping task is identifying the grasp itself (within the static grasp phase). The proposed strategy of identifying the grasp is to map the low-level hand configuration to increasingly more abstract grasp descriptions. The abstract grasp descriptions are useful because they are manipulator-independent.

引用

页码：432 / 443