Project Title:

Kinematic Path Planning for Redundant Manipulators using a Path Space Approach

  94-1    05.03     5355      



 Kinematic Path Planning for Redundant Manipulators using a Path

Space Approach

Abstract:

Intelligent robotic systems for future space exploration will

rely heavily on. autonomous and semi- autonomous motion planning

for task completion. Communication time delays render

teleoperation of manipulators infeasible over the large distances

expected between operators and remote sites. We propose a general

and versatile kinematic path planner using a novel path space

approach. The planning problem, posed as a finite-time nonlinear

control problem, is transformed into a static root- finding

problem and iteratively solved. Optimality can be incorporated as

a secondary criterion. Compared to existing approaches, this

method is less prone to problems such as arm singularities and

local minima of potential fields. Unlike many search-based

methods, it does not rely on the construction of a collision-free

search space, which can become prohibitively large for many

degrees-of- freedom. In contrast with existing autonomous

techniques, our approach is designed to be compatible with a

supervisory layer, either human or artificial intelligence based.

Phase I will develop this technique to solve current and future

robotic path planning scenarios for space missions, such as ORU

changeout tasks, orbiter tile inspection, and space structure

assembly and maintenance. 


The anticipated result of this effort is a general and versatile

path planner, applicable to a wide range of robotic tasks, such

as welding/assembly of automobile structures, automated

manufacturing systems, material handling in hazardous

environments such as nuclear site cleanup, undersea operations,

etc. Our approach also addresses the motion planning problem for

non-holonomic systems. Potential applications include articulated

wheeled vehicles (eg. tractors with multiple trailers). This

technology will be invaluable to the development of intelligent

vehicle systems for the emerging IVHS market. 


Key Words


Scientific Systems Company, Inc.

500 West Cummings Park, Suite 3950

Woburn, MA 01801-6336