Project Title:

Fast Generation of NURBS Surfaces from Polygonal Mesh Models of Human Anatomy

Principal Investigator:

Charles Anderson, Associate Professor
Department of Computer Science
Colorado State University

Collaborating Companies:

Visible Productions

Company Representative:

Thomas McCracken, President

Visible Productions, Inc., of Fort Collins, CO, produces three-dimensional human models that are recognized as some of the most accurate models in the world. Their models currently are based on meshes of three-dimensional triangles. Such meshes can be rendered as smooth surfaces by interpolating color values across a triangular mesh, but for a number of applications the smooth surface must be explicitly represented. Clients for Visible Productions' models have asked for surfaces defined by NURBS (Non-Uniform Rational B-Splines). This project will develop and implement algorithms for transforming polygonal meshes into NURBS. This will probably require a time-intensive, interative optimization process. We will investigate the use of neural networks to by-pass a large part of the optimization process.

Project Title:

Investigation of Use of MPI/RT to Re-Engineer Real-Time Distributed Flight Simulation

Principal Investigator:

Robert G. Babb, II, Associate Professor
Department of Mathematics and Computer Science
University of Denver

Collaborating Companies:

SymSystems, LLC

Company Representative:

Dave Flanagan, Vice President

Currently, real-time application developers must use platform-specific communication and coordination mechanisms, and experiment to achieve desired real-time characteristics. MPI/RT is an emerging standard whose goal is to provide middleware for programmers to create distributed applications with real-time performance portability. A variety of real-time design paradigms are supported explicitly: time-driven, event-driven, and priority-driven, as well as mixtures of these models. This project will investigate use of MPI/RT mechanisms to re-engineer SymSystems' Mission Simulator System (MSS). Primary goals for this re-engineering are to 1) apply and exercise the MPI/RT mechanisms in a typical real-time commercial application. 2) increase future architecture-independence of MSS by making explicit and standardizing an API for user-produced extensions to MSS 3) use MPI/RT's design paradigms to make explicit the various real-time paradigms used by the current MSS software architecture.

Project Title:

WEB VHR Retrieval Technology

Principal Investigator:

Dushan Badal, Professor
Division of Computer Science
University of Colorado - Colorado Springs

Collaborating Company:

Exatech Corporation

Company Representative:

George Fredericks, President

We propose to investigate novel VHR (Very Highly Relevant) retrieval technology which will avoid information overload being experienced by web users. The problem is that everybody has browsers and no time to browse even a small part of the potentially useful information that is freely available and growing unimaginably. We propose to investigate much more targeted semantics based search techniques for web pages or documents so that the user receives only text objects which are very highly relevant to the subject at hand. In order to support such VHR retrieval, we propose to investigate the query-by-example (qbe) paradigm whereby the user can ask "find me more documents like this one" rather then what is currently done "find documents which have these keywords connected by and, or, not". We also propose to investigate term proximity based ranking rather than the currently used term frequency based ranking.

Project Title:

High Speed Onboard Radar Processor

Principal Investigator:

Tamal Bose, Associate Professor
Department of Electrical and Engineering
University of Colorado - Denver

Collaborating Company:

Data Fusion Corporation

Company Representative:

Woody Kober,

This project deals with the development and implementation of a fast algorithm for processing data from Synthetic Aperture Radars (SAR).

In recent years, SAR has evolved to satisfy a variety of civilian and military uses. These applications center on terrain mapping and target imaging. The raw data collected by a SAR must be processed in order to form the image. This is done by the Image Formation Processor (IFP). In order to form the image, the IFP performs several functions such as motion compensation, range deskewing and the Range Migration Algorithm (RMA). The RMA is a computationally intensive and time-consuming algorithm and therefore cannot be performed in real-time on board the aircraft carrying the SAR. The goal of this project is to modify the RMA in order to drastically reduce its computational complexity so that we can implement the entire algorithm in a single digital signal processing (DSP) chip operating in conjunction with a personal computer (PC). The entire system can then be carried on board the aircraft for real-time processing.

Project Title:

WINPLAN: Wireless Information Network Planning

Principal Investigator:

Edward Chow, Associate Professor
Division of Computer Science
University of Colorado - Colorado Springs

Collaborating Company:

Omnipoint Technologies, Inc

Company Representative:

Dean Angelico, Senior Manager

WINPLAN: Wireless Information Network Planning.
The proposed research deals with the development of efficient algorithms and simulation tools for Wireless Information Network Planning and Management. A software design system called Wireless Information Network Planning (WINPLAN) will be built to facilitate the design and evolution of wireless information networks with realistic user traffic modeling, efficient dynamic channel assignment algorithms, and flexible cell layout tool. The WINPLAN system will assist the network administrators in their network planning and management tasks. It will facilitate the network designers to improve the network efficiency and reliability.

Project Title:

Measurement of Normal Knee Kinematics from X-ray Images

Principal Investigator:

William A. Hoff, Assistant Professor
Division of Engineering
Colorado School of Mines

Collaborating Company:

Rose Musculoskeletal Research Laboratory

Company Representative:

Richard Komistek, Director of Development

The objective of this project is to develop a technique for measuring the kinematics of normal knees from X-ray fluoroscopy images, "in vivo". The bones of the human knee joint (femur, tibia, and patella) undergo full six degree-of-freedom (DOF) motion (motion about 3 rotational axes and along 3 translational axes) during normal flexion. Understanding this motion is critical for designing implants which better replicate normal knee kinematics. It is important to measure the motion in actual patients performing weight bearing activities, to take into account the effect of the muscles and ligaments. Previous methods for measuring normal knee kinematics "in vivo" have been restricted to measuring motion within the plane of the X-ray image (one rotational and two translational degrees of freedom). Over the past three years, the P.I. and his collaborators at the Rose Musculoskeletal Research Laboratory (RMRL) have developed algorithms and software to measure full 6 DOF motion of artificial knee implants from X-ray fluoroscopy images. This system has been highly successful, and has been extensively used to analyze hundreds of patients and dozens of implant models. The objective of the proposed research is to extend this system to analyze normal knees.

Project Title:

Precise Control of Elastic Components for Tape Systems

Principal Investigators:

Lucy Pao, Assistant Professor
Department of Electrical and Computer Engineering
University of Colorado - Boulder

Collaborating Company:

Storage Technology

Company Representative:

John Mantey, Manager, Nearline Servo Systems

Flexible structures are prevalent in manipulators, machine tools, measurement systems, space structures, and many other systems. We propose to develop and explore methods of designing actuator commands to allow for rapid and robust control of flexible systems in the presence of parameter uncertainty. The methods we will develop will exploit the notions of input shaping to provide guarantees of desired levels of robustness. Previously proposed techniques have primarily focused only on providing robustness to uncertainty in the natural frequencies and are often fairly sensitive to variations in damping. Because damping is often especially difficult to model accurately, we shall study and develop command shaping approaches for flexible systems that are robust to both frequency and damping uncertainty. We will further investigate techniques for making our algorithms adaptive, so that the actual parameter values for each system can be estimated and used in fine tuning the control algorithms on-line. Finally, we will work with researchers and engineers at Storage Tek to demonstrate the developed algorithms by applying them to tape systems.

Project Title:

NON-PUSH: A Program for Nonlinear Push-Over Analysis of Building Structures

Principal Investigator:

Enrico Spacone, Assistant Professor
Department of Civil, Architectural and Environmental Engineering
University of Colorado - Boulder

Collaborating Companies:

KL & A of Colorado

Company Representative:

Gregory Kingsley, Principal

This project proposes the development of NON-PUSH, a PC-based computer program for the nonlinear push-over analysis of reinforced concrete and steel frame structures. The program will link two new libraries, one of frame elements and the another of material constitutive laws, to an existing nonlinear finite element program. A user-program interface will be added to the program to ease the input data task. The program is the first step of a long-term effort to fill the gap between the need to routinely perform nonlinear frame analyses, as prescribed by the new generation of seismic building codes, and the almost total lack of efficient, user-friendly and well documented nonlinear frame analysis programs.

Project Title:

Ultrasonic Damage simulation in Engineering Materials

Principal Investigator:

Kaspar Willam, Professor
Department of Civil, Architectural and Environmental Engineering
University of Colorado - Boulder

Collaborating Companies:

APTEK Incorporated

Company Representative:

Yvonne Murray, Project Manager

We propose to develop a graphics-oriented simulation package for quantitative ultrasonics. In short an "Ultrasonic Damage Simulator", to evaluate material degradation in terms of propagative material properties. Examples of ultrasonic damage experiments include damage assessment of civil and aerospace structures, where stiffness degradation provides an indicator of the health and life cycle performance of structural components and structural systems.

To this end, the " Ultrasonic Damage Simulator" will evaluate and graphically display phase velocities diagrams (slowness surfaces) in engineering materials that are deteriorating due to mechanical and environmental loadings. Along this line, the driver will establish a testbed to interrogate in a quantitative manner constitutive relations for a wide variety of material formulations with regard to their propagative properties.

The software will be developed on a SUN platform and will be written in Fortran 90 and in C++ using Open GL/Open Inventor for visualization and animation. In the second phase of the project, we intend to download the "Ultrasonic Simulator" to high end PC's that are widely used in engineering practice and classroom environments.