Medical telemetry and signal processing can be used to monitor and process patients’ physiological parameters over a distance via radio frequency (RF) communications between a patient’s transmitter and a central monitoring station. This technology delivers mobility, comfort, and higher levels of out-patient care since the devices are less cumbersome than traditional equipment. Development of an intelligent wireless cardiac signal recording/processing technology is vital for continuous monitoring and diagnosis of cardiac patients. Such technology will provide the ability to remotely monitor patients and determine cardiac patterns before a cardiac attack occurs. The objective of this research is to design an intelligent cardiac monitoring system which is based on wireless skin electrodes. Each skin electrode consists of an ultra-miniaturized extremely low-power telemetry unit which transmits the cardiac signal to the central unit. The central unit collects the signals from all wireless electrodes and transfers the information to a processing unit to detect any cardiac arrhythmia using intelligent signal processing. The specific aims of this research are:

1) Explore techniques for optimal design of ultra-low power transceivers, sensor interfaces and miniaturized antennas;

 2) Propose alternative approaches for system integration and packaging;

3) Devise enhanced cardiac signal processing algorithms and;

 4) Perform tests on patients to study cardiac patterns by gender and age.

Lead Principal Investigator/Coordinator

Kamran Entesari
Professor

Texas A&M University

Principal Investigators

Reza Tafreshi
Associate Professor

Texas A&M University at Qatar

 

Edgar Sanchez Sinencio
Texas A&M University

A new fueling control design tool is sought to produce fuel efficient engines with zero tailpipe emissions. The broad objectives of this work include development of an automated nonlinear dynamic system modeling process applicable to systems with delays, a linear parameter varying (LPV) control design strategy integrated with an adaptive pre-filter to maximize the closed-loop performance, and a nonlinear feedforward controller leading to rapid tracking control. To achieve the proposed goals, we develop

(i) a nonlinear modeling process based on a functional power series to address time delay systems with input nonlinearity and remove the guess-work in selecting model nonlinearity through System Probing;

(ii) a method for selecting performance weights associated with H_inf synthesis problem for the class of nonlinear systems considered in the project to satisfy the prescribed output performance despite the presence of disturbance inputs;

(iii) a synthesis process in designing a feedforward compensator based on an inversion of functional power series;

(iv) an LPV control design methodology for the engine fueling system in the presence of time-varying delays. A prefilter is designed to maximize the closed-loop bandwidth regardless of engine speed variations. The proposed design methods will be experimentally validated using the UH state-of-the-art engine facility. Remote testing and data collection are used by the collaborating institutions to access the experimental facility.

The proposed research is focused on nonlinear structural health monitoring (NSHM) where linear models are incapable of capturing the dynamic response of the system. New engineering tools will be created to advance the NSHM knowledge base by harvesting the Volterra series (VS) and artificial neural networks (ANN) knowledge bases. A transformation of an ANN into a VS will be performed for a meaningful interpretation of the internal workings of an ANN. Conditions will be developed that conclusively determine if an ANN has been successfully trained. We will experimentally verify our findings by constructing two scale-model buildings. One building will be used to develop methods to quantify the effects of variable operating conditions (different operating conditions due to loading levels and nonlinearity excitement). The second building will have several forms of repairable damage built in. Our NSHM will be accomplished through adaptive ANN modular (per floor) models. This modeling approach is the most compact and allows one to directly isolate changes to specific floors. Accuracy issues of NSHM will be addressed from a rigorous statistical uncertainty study centered on quantifying environmentally variability. The four major tasks are:

(1) Development of an automated online ANN training tool

(2) Development of a VS ANN Interrogation Tool

(3) Development of Structural Health Estimation Methods

(4) Design and Fabrication, Environmental Bounding and Experimental Validation

Lead Principal Investigator/Coordinator

The proposed research is within an extended notion of what is known as "Smart Field" which incorporates the usage of “smart” robotic coordination to enhance the overall system performance and safety in a field. The focus of the project is on developing methods for safe coordination of groups of multiple vehicles accomplishing multiple tasks in the presence of a variety of physical, collision avoidance, and informational constraints. Safe coordination implies the accomplishment of the vehicles’ individual and group goals while never violating the overall multi-vehicle system safety. Coordination of multiple vehicles is considered safe if there is a guarantee of no collisions between the vehicles and between the vehicles and static objects. The individual and group goals are defined based on numerous applications including surveillance and reconnaissance, control of mobile sensor networks, manipulation of static objects using mobile robots, and air and highway traffic control, to name a few. The main deliverable of this project is to provide methodologies that would lead to implementable control laws for coordination of a variety of vehicles in the presence of safety, motion, and communication constraints. The theoretical results will be tested on a number of experimental vehicle platforms which involve very minimal human and machine interaction.

Lead Principal Investigator/Coordinator

Principal Investigators

High cost and downtime time caused by drillstring failures can seriously maim productivity of oil and gas drilling companies. These failures are caused by nonlinear vibrations, buckling, fatigue, and wear. The objective of this research is to mitigate these failures via

(1) improved, measurement based models of interaction forces that act between rotating assembly and hole/fluid/formation/casing,

(2) improved drillstring vibration simulation,

(3) mitigation of drill bit, BHA, and drill collar vibration by using effective protocols for drillstring operators,

(4) development of state-of-the-art monitoring system based on vibro-acoustic transducers and wired relays, and

(5) education of future and practicing drilling engineers. We will build 3 test rigs and develop finite element model of the system. The TAMU-Qatar rigs will provide experimental data to characterize fluid interaction as well as bit interaction. The TAMU deep-hole rig in College Station will be used to verify drillstring simulation software and to perform advanced testing of vibration suppression and measurement concepts. The TAMU rig will make use of an existing facility at Riverside Campus to reduce cost. Unique contributions will be realistic test rigs, use of advanced active vibration control for the BHA, accurate interaction force measurement and modeling, and real-time monitoring of drillstring dynamics by using vibro-acoustic measurement system.

Lead Principal Investigator/Coordinator

Principal Investigators

Failure of oil/gas pipelines could be very costly financially and ecologically. A good portion of oil/gas companies’ maintenance budget is spent on inspection of pipelines internally and externally. Internal inspection is carried out through pigging, whereas the external inspection is still primitive and rare and usually carried out through visual inspection. It is proposed here to design an autonomous robotic system for the external inspection of oil/gas pipelines consisting of a rugged mobile robotic platform outfitted with multi-physics sensory devices. The sensory systems will use state of the art acoustic and magnetic flux leakage (MFL) techniques to automatically scan and detect pipeline defects such as cracks, dents, and gouges. The MFL sensing consists of creating a magnetic field strong enough to saturate the pipe and having a ring of Hall Effect sensors to pick up field leakages. Using advanced inverse problem techniques, the “types” and “sizes” of pipe defects will be determined. The proposed acoustic sensory system is to identify the “locations” and “shapes” of structural defects by emitting a tuned ultrasonic wave and then measuring its reflected waves from the structural defects by using an array of ultrasonic transducers. Since the tuned ultrasonic wave can propagate long distances with small spatial decay rates, the proposed procedure allows inspection of a long pipeline with a relatively small number of scanning locations.

Lead Principal Investigator/Coordinator

Principal Investigators