This page is provided for historical and archival purposes only. While the seminar dates are correct, we offer no guarantee of informational accuracy or link validity. Contact information for the speakers, hosts and seminar committee are certainly out of date.
The Robotics Institute Carnegie Mellon University
An Intelligent Manufacturing Workstation is a planning system, controller and a machine tool that acts as a standalone design-to-production system. Our group has built two such workstation: one for machining and now one of bending. This talk will explore the architecture of the Intelligent Bending Workstation (IBW), concepts of process planning and sensor based control that make it feasible.
The IBW is comprised of a parallel designer that maintains 1-to-1 relationship between the 2D and 3D part topologies, a multi-agent planning system (bend sequence planner, holding expert, tooling expert and motion expert), and a real-time sensor-based control system. All of these systems communicate in FEL (the Feature Exchange Language), which is an extensible language for multi-agent communication. We will discuss the algorithms used by these agents, we will illustrate a new algebraic scheme that should significantly speedup other process planning tasks, and we will show how learning is used to improve the process from run to run.
Using industrial test parts, we will show how the IBW plans, sets up and produces bent sheet metal parts. In one example, it took 22 minutes to produce a part that took human experts 3 days to make in the past. This result encourages us that it will be possible to achieve economic ordering quantities of one. In closing, we will speculate about what is possible given the existence of Intelligent Manufacturing Workstations.
Host: Yangsheng Xu (firstname.lastname@example.org) Appointment: Lalit Katragadda (email@example.com)