Supervisory control for automated assembly based on subassembly manipulation
Abstract
This thesis describes Motivator, a task planner for the Automated Assembly Workcell. Motivator takes information from a CAD data-base and vision system to determine the optimal sequence of assembly and uses this plan to produce the trajectories required for the assembly process. An optimal assembly sequence necessarily must consider subassemblies, their manipulation, their graspability, and the ability to mate them. The thesis concentrates on the manipulability of subassemblies; that is, the determination of which subassemblies in an assembly can be transported without disassembly occurring. A new methodology for the determination of stability of subassemblies is presented. This method is based on the geometries of the parts within the subassembly and friction; it does not account for the inertial forces of motion, but this approximation is decidedly useful. The stability graph and an algorithm to propagate constraints within a subassembly are introduced to define and quantify the manipulability of objects. Results indicate that this method is faster than methods which utilize the Simplex algorithm, and is more general than rule based methods.
Degree
Ph.D.
Advisors
Shoureshi, Purdue University.
Subject Area
Mechanical engineering
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.