Witold Jacak

CAST Tools for Intelligent Control in Manufacturing Automation

A comprehensive framework for design of an intelligent cell-controller requires integration of several layers of support methods and tools. We have proposed an architecture that facilitates an automatic generation of different plans of sequencing operations, synthesis of action plan for robots servicing the devices, synthesis of the workcells simulation model, and verification of control variants based on simulation of the overall cells architecture. The real-time discrete event simulator is used next to generate a sequence of future events of the virtual cell in a given time-window. These events are compared with current states of the real cell and are used to predict motion commands of robots and to monitor the process flow. The architecture, called Computer Assisted Workcell, offers support methods and tools at the following layers of control: organization, coordination, and execution.

Simulation-based planning of robot tasks in flexible manufacturing

A framework is proposed for support of design, task planning, and simulation of automated manufacturing systems. The framework establishes a hierarchy of method banks essential for improving the efficiency and cost effectiveness of manufacturing processes. The methods should support automatic generation of sequencing rules, design and configuration of the manufacturing facility and equipment, synthesis of task oriented robot programs, and generation and execution of simulation models of a manufacturing system. In this paper, each layer is addressed and preliminary results that apply simulation to interpret and test task oriented robot programs are discussed.<<ETX>>

Hybrid Evolutionary Programming: the Tools for CAST

With the development of new computing paradigms, such as neural networks and genetic algorithms, new tools have become available in computer-aided systems theory. These tools can be used to tackle problems that are considered “hard” in traditional systems theory, like the modeling and identification of nonlinear dynamical systems.

Model-based workcell task planning and control

HANS L. STEINMETZ is a native of Germany, where he obtained a degree in electrical engineering from the T. H. Aachen. After working as an application engineer for Brown, Boveri and Company in Mannheim, Germany, Mr. Steinmetz came to the United States, where he worked for the Allis-Chalmers Manufacturing Corporation in Milwaukee, Wisconsin on the design of industrial motor control systems. During this period he obtained his MSEE from the University of Wisconsin. He then joined the California Research Corporation in Richmond, California where he was concerned with the simulation of chemical processes. During this time he taught courses in analog and hybrid computer programming at the University of California at Berkeley. After a brief excursion into the field of oceanography at the Chevron Oil Field Research Company in La Habra, California he joined the Orlando, Florida Division of the Martin Marietta Corporation. Here his interests center about the application of statistical methods to missile flight trajectory analysis using hybrid computers.

Lifelong Learning Approach to Intelligent Agents Modeling

In this paper, we presented an application of neural network-based models in the intelligent agent domain. The use of neural networks has the advantage that the model can adapt itself to changing environment conditions by simple retraining steps. This is illustrated by two functions needed for the lifelong learning-based modeling of an intelligent robotic agent: The function for generalizing sensor observations to conceptual states, and the function for modeling the effects of robot actions on the conceptual state space. An example shows that the algorithms can be applied in real-world environments.

Neural network-based modeling of robot action effects in conceptual state space of real world

This paper gives the concept of an autonomous robotic agent that is capable of showing both machine learning and reactive behavior. The first methodology is used to collect information about the environment and to plan robot actions based on this information while the robot is performing tasks. Processing and storing information obtained during several task executions is called lifelong learning. Reactive behavior, the second desirable feature of an autonomous robot, is needed to execute the actions in a dynamically changing environment. This paper presents the main components of the machine learning part of the autonomous robotic agent: the conceptual state generalization module that forms conceptual states as categories (clusters) of observation vectors obtained from the sensor system; and the robot behavior effects modeller that calculates the next conceptual state from the current state and an action obtained from the action planner.

Using simulation in process engineering

The flexible and economic production of goods requires a new level of automation. Workcells integrating manufacturing stations and robots, form the basis of a flexible manufacturing process. The manufacturing process to be performed in such a cell also needs definition. A prerequisite is an adequate process definition environment which uses computer support to define the manufacturing process. The basis of such an environment is sketched here.<<ETX>>

Towards Design and Control of High Autonomy Manufacturing Systems

In this paper, requirements for design of high autonomy manufacturing systems are stipulated. Effods towards amalgamating the autonomous architecture and its real world component (i.e., a flexible manufacturing system) are presented. Planning and control principles derived from discrete event modeling techniques are summarized.

Software Processes, Work Flow and Work Cell Design - Separated by a Common Paradigm?

This paper discusses similarities of computer support in three areas: development of software, supporting the flow of work/documents in an office and the production of goods in a flexible work cell. All three areas obey the same underlying principle, i.e. defining a process and then interpreting the process by an enactment (interpretation) mechanism. In the paper elementary questions with respect to the similarity are raised and tentative answers given. The conclusion is that the use of the same paradigm is valid, that considerable difference have to be considered in its implementation.