GGrantIndex
← Search

CAREER: Nonprehensile Mobile Manipulation

$280,941FY2000CSENSF

Rensselaer Polytechnic Institute, Troy NY

Investigators

Abstract

This is the first year funding of a four year continuing award. This proposal describes an integrated career development plan with research and educational activities centered on methods for mobile robot systems to display a basic competence in manipulation. The approach advanced by this proposal is to equip nonholonomic mobile robots with very simple manipulators and use nonprehensile manipulation techniques to accomplish a variety of manipulation tasks. We will initially focus on two types of manipulators: a "plan" manipulator and a "finger" manipulator. A single robot with one of these manipulators can push or roll an object along the ground. Although this seems limited, a wide variety of objects can be manipulated in this manner. With two or more such robots, a much greater variety of manipulation is possible: objects can be lifted, reoriented, carried, and so on. The proposed research has three main components: generalizing existing nonprehensile manipulation strategies and developing new ones for mobile manipulators with nonholonomic motion constraints; creating a unified framework for manipulation planning that can reason about different nonprehensile manipulation strategies; and the distribution of intelligence among multiple robots in order to achieve robust cooperative manipulation. The proposed educational activities complement the research plan. The research will require custom-built mobile manipulators and the development of nonprehensile skills. These activities are excellent opportunities for hands-on education, and they are well within the scope of class projects and independent studies. There are three parts to the proposed educational activities: engagement of undergraduates in research; development of a hands-on course covering manipulation, mobile robotics, and robot-design; and writing a text to serve as a reference for robot design and programming.

View original record on NSF Award Search →