A LOW-LEVEL COLLISION AVOIDANCE OPERATION (ROBOT, PATH-PLANNING, MANIPULATOR, AUTOMATIC GUIDED VEHICLE)

RUSSELL KENT READ, Purdue University

Abstract

Industrial automation, while becoming more widespread, suffers a notable shortcoming. Lack of flexible operation both restricts the applicability of robotic or automation techniques and requires significant expenditures for conveyors, parts handlers, safety screens, and other devices which ensure a safe, structured environment. The use of path planning in conjunction with collision avoidance can provide the necessary flexibility. In particular, a low-level operation is described which determines collision-free robot movement in the presence of 3-dimensional moving objects. A commercial vision system is employed to obtain a description of objects in a scene at each image sampling point. A matching algorithm permits trajectory estimation for the objects. A goal-seeking Next Move is iteratively generated for the case of a point robot or a multi-link robot. Several imminent-collision recovery methods are discussed.

Degree

Ph.D.

Subject Area

Electrical engineering

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS