3D Stereo with Applications to Mine Safety

Project Team: John P. H. Steele, Mark Whitehorn, Chris Debrunner, and Tyrone Vincent

This project is focused on developing a real-time 3D stereo imaging system for modeling mining environments. Being able to "see" the world and develop a model of it is fundamental to a number of robotic applications in mining.

Advancing Health Monitoring Techniques for Infrastructure and Intelligent Construction

Project Team: Mike Mooney, Karen Furlani
Sponsors: National Science Foundation

The research focuses on infrastructure health monitoring issues, i.e., developing and deploying techniques to provide comprehensive, real time monitoring of geotechnical and structural systems. Research activity involves the design and exploration of wired and wireless sensor networks to monitor performance (deformation, stress) of buried structures (pipelines, tunnels), adjacent buildings, retaining wall structures, and the soil. Research activity also involves exploring sensor/medium compatibility, optimization of sensor placement on subsurface and surface structures, sensor durability, and model based/non-model based system identification techniques to assess structural performance from collected data. Finally, the research will explore issues pertaining to the integration of health monitoring with ‘on-the-fly’ design and with construction operations, e.g., web-based presentation of construction field data; the use of health monitoring data to control wall anchors, excavation, brace actuation, etc; and health monitoring data within a decision support system.

Control of CIGS Deposition Process

Project Team: Matt Hilt, Jeff Hanna, Tyrone Vincent, JP Delplanque, Robert Kee
Sponsors: ITN Energy Systems

In order to successfully produce photovoltaic modules the operating conditions for the thin film deposition processes must be selected, and feedback control (at some level) must be implemented in order to achieve these operating conditions. This project involves model development of the deposition equipment to assist both of these efforts. An integrated equipment model constitutes a virtual manufacturing capability which allows the exploration of operating conditions with the cost of only computation. At particular operating conditions, the dominant dynamic behavior can be extracted, allowing the development of real time feedback control.

Intelligent Vibratory Roller Compaction

Project Team: Mike Mooney, Paul Van Susante
Sponsors: National Science Foundation, Ingersoll-Rand Corporation

This research aims to transform traditional vibratory earthwork compaction into an intelligent process whereby the machine automatically adjusts vibration frequency and amplitude to optimize the soil compaction process. The project involves: (1) developing a sensor network to characterize coupled machine and soil behavior; (2) developing the relationships between sensed machine vibration and the soil properties throughout compaction; and (3) developing the feedback control strategies to vary vibration frequency and amplitude.

Model Free Fault Detection for Nonlinear Systems

Project Team: Renee Spinhirne, Tyrone Vincent
Sponsors: National Science Foundation

Fault detection of linear systems is by now well established, but application of these techniques to nonlinear systems can lead to a loss of performance. The market for fault detection is growing, as more and more manufacturing and industrial centers are interested in improving their productivity by reducing unplanned maintenance and limiting losses to product due to equipment failure. In addition, many manufacturers have installed sophisticated sensor and data collection networks in order to closely monitor their processes, which would enable the application of very sophisticated fault detection schemes. This project is aimed at developing a theory of the interactions between the system identification and fault detection processes. The focus is on a general, practical method of nonlinear fault detection which is easily implemented on input/output models that are generated by modern nonlinear system identification methods.

Smart Bits - Sensing at the Machine-Rock Interface

Project Team: John P. H. Steele, M. Ugur Ozbay

This project is developing new sensing technology so that mining machines can sense what's happening as they cut through rock. This information will be used to optimize their performance and will lead to more autonomous mining operations.

Vision Based Control of Droplet Manufacturing and Welding

Project Team: Tyrone Vincent, Chris Debrunner, John Steele, JP Delplanque
Sponsors: National Science Foundation

Computer vision has already found significant use in manufacturing processes for activities such as quality control. However, the use of vision in order to control a dynamic system pushes the requirements of these sensors. Recently, the combination of fast computing and high density imaging arrays has made computer vision a valid sensor for the control of dynamic systems, and this makes the use of cameras in robotics, welding, and droplet-based manufacturing possible. Our focus will be on integrating the rich sensor information available from vision with reduced order fluid-thermal dynamic models for the purposes of real time control.

Mobile Ad Hoc Networks (MANET)

Project Team: Tracy Camp, William Navidi, Michael Colagrosso, Jeff Boleng, Stuart Kurkowski, Peiling Yao

Sponsor: National Science Foundation

Ad hoc networking involves computers, typically wireless mobile nodes, that cooperatively form a network without specific user administration or configuration.  In other words, ad hoc networking allows an arbitrary collection of mobile nodes to create a network on demand. There are numerous scenarios that do not have an available network infrastructure and could benefit from the creation of an ad hoc network. For example, a rapid installation of a communication infrastructure during a natural/environmental disaster (or a disaster due to terrorism) that demolished the previous communication infrastructure. Our focus is the investigation of network layer protocols. Currently, five projects are underway. For further details, see the Toilers webpage.