AMP-CAD: Automatic assembly motion planning using CAD models of parts
Abstract
A common problem in robotic assembly is that of mating tightly fitting parts when the locations and the dimensions of the parts are somewhat uncertain. We will present a planning system that develops motion plans for assembly mating operations which must be carried out under uncertain conditions; these plans are made up of point-to-point motions and force/torque guided motions. Our planner uses graph search over a potential field representation, derived from the CAD models of the parts, to calculate candidate assembly paths. Given the tolerances of the parts and other uncertainties, these paths are then analyzed for the likelihood of collisions. Those path segments that are prone to collisions are then marked for execution under force/torque control. The calculation of the various motions is facilitated by an object-oriented and feature-based assembly representation which was designed to aid in the creation of a potential field representation used in planning and in the incorporation of various uncertainties into this potential field representation. An integral part of the planning system is the execution unit. Residing in this unit is knowledge of the different types of error detection and recovery strategies. During the execution of a motion, this unit invokes the error detection and recovery strategy appropriate to the geometric constraints relevant to the motion. This system, called AMP-CAD, has been experimentally verified using a Cincinnati Milacron T$\sp{3-}$726 robot and a Puma 762 robot on a variety of assemblies.
Degree
Ph.D.
Advisors
Kak, Purdue University.
Subject Area
Mechanical engineering|Electrical engineering|Computer science
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.
