Dana Hughes

Dana Hughes
University of Colorado Boulder | CUB · Department of Computer Science (CS)

BSEE, MSEE, MSCS, PhD

About

30
Publications
10,687
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
420
Citations
Introduction
Dana Hughes is a postdoctoral fellow at the Robotics Institute at Carnegie Mellon University. He received his PhD in Computer Science from the University of Colorado at Boulder, a MS in Computer Science at the College of Charleston, a MS in Electrical Engineering at the University of Missouri---Rolla, and a BS in Electrical and Computer Engineering at Colorado State University. His research focuses primarily on learning behavior for collaborative robots, robotic materials, novel methods for sensing and perception, and AI and machine learning applications.
Additional affiliations
January 2010 - May 2012
College of Charleston
Position
  • Research Assistant
January 2001 - May 2003
University of Missouri
Position
  • Research Assistant
January 1999 - December 2000
Colorado State University
Position
  • Undergraduate Research Assistant

Publications

Publications (30)
Preprint
Full-text available
We present a library to automatically embed signal processing and neural network predictions into the material robots are made of. Deep and shallow neural network models are first trained offline using state-of-the-art machine learning tools and then transferred onto general purpose microcontrollers that are co-located with a robot's sensors and ac...
Conference Paper
Full-text available
We describe a self-sensing composite capable of making online estimations on the location of an impact within its structure without the need for wires or offline data processing. We embed a network of piezoelectric sensors in the material during manufacturing and connect each sensor to a dedicated micro-controller and power source. Triangulation is...
Conference Paper
Full-text available
JythonMusic is a software environment for developing interactive musical experiences and systems. It is based on jMusic, a software environment for computer-assisted composition, which was extended within the last decade into a more comprehensive framework providing composers and software developers with libraries for music making, image manipulati...
Chapter
We investigate distributing convolutional neural networks (CNNs) for human activity recognition across computing nodes collocated with sensors at specific regions (body, arms and legs) on the wearer. We compare four CNN architectures. A distributed CNN is implemented on a network of Intel Edison nodes, demonstrating the capability of performing rea...
Article
We describe a flexible robotic skin module that can measure proximity, contact and force, as well as algorithms for detecting obstacles, human hands, and affective touch gestures. The computational requirements of the proposed architecture are in line with the capabilities of a small microcontroller, allowing for a majority of the data to be proces...
Article
Full-text available
We describe a self-sensing composite capable of making online estimations on the location of an impact within its structure without the need for wires or offline data processing. We embed a network of piezoelectric sensors in the material during manufacturing and connect each sensor to a dedicated micro-controller and power source. Triangulation is...
Article
Full-text available
We present an radio-frequency (RF)-based approach to gesture detection and recognition, using e-textile versions of common transmission lines used in microwave circuits. This approach allows for easy fabrication of input swatches that can detect a continuum of finger positions and similarly basic gestures, using a single measurement line. We demons...
Article
Full-text available
This paper reviews machine learning applications and approaches to detection, classification and control of intelligent materials and structures with embedded distributed computation elements. The purpose of this survey is to identify desired tasks to be performed in each type of material or structure (e.g., damage detection in composites), identif...
Conference Paper
While several sensing modalities and transduction approaches have been developed for tactile sensing in robotic skins, there has been much less work towards extracting features for or identifying high-level gestures performed on the skin. In this paper, we investigate using deep neural networks with hidden Markov models (DNN-HMMs), geometric moment...
Article
Full-text available
We present a soft robotic skin that can recognize and localize texture using a distributed set of sensors and computational elements that are inspired by the Pacinian corpuscle, the fast adapting, uniformly spaced mechanoreceptor with a wide receptive field, which is responsive to vibrations due to rubbing or slip on the skin. Tactile sensing and t...
Conference Paper
Full-text available
We present SwitchBack, a novel e-textile input device that can register multiple forms of input (tapping and bidirectional swiping) with minimal calibration. The technique is based on measuring the input impedance of a 7 cm microstrip short-circuit stub consisting of a strip of conductive fabric separated from a conductive fabric ground plane (also...
Conference Paper
Full-text available
We present SwitchBack, a novel e-textile input device that can register multiple forms of input (tapping and bi-directional swiping) with minimal calibration. The technique is based on measuring the input impedance of a 7 cm microstrip short-circuit stub consisting of a strip of conductive fabric separated from a conductive fabric ground plane (als...
Conference Paper
We present a soft, amorphous skin that can sense and localize textures. The skin consists of a series of sensing and computing elements that are networked with their local neighbors and mimic the function of the Pacinian corpuscle in human skin. Each sensor node samples a vibration signal at 1KHz, transforms the signal into the frequency domain, an...
Article
Monterey Mirror is an experiment in interactive music performance. It is engages a human (the performer) and a computer (the mirror) in a game of playing, listening, and exchanging musical ideas. The computer side involves an interactive stochastic music generator which incorporates Markov models, genetic algorithms, and power-law metrics. This app...
Conference Paper
Full-text available
We present an automated approach to music search and playlist generation based on fractal dimensions of music. We compute 372 power-law metrics per song capturing statistical proportions of musical material. Using attribute selection and principal component analysis, we have reduced these metrics to approximately 45 independent features. These have...
Article
The use of combined embedded modulated scattering technique and near-field microwave nondestructive testing techniques is investigated as a novel method for evaluating the dielectric properties of a material. The forward formulation for determining the reflection coefficient at the aperture of a waveguide radiating into a dielectric half-space in w...
Article
Full-text available
The use of combined embedded modulated scattering technique and near-field microwave nondestructive testing techniques is investigated as a novel method for evaluating the dielectric properties of a material. The forward formulation for determining the reflection coefficient at the aperture of a waveguide radiating into a dielectric half-space in w...
Article
Full-text available
The use of carbon fiber reinforced polymer (CFRP) composites has been investigated in the past few years as a means to rehabilitate aging concrete structures. The detection of disbonds and delaminations between CFRP laminate and the concrete substrate is a critical issue in this area, as the presence of such defects adversely affects the effectiven...
Conference Paper
Full-text available
Evaluation of thickness and material properties of coatings and dielectric slabs is an important practical issue. Microwave nondestructive testing techniques, using open-ended rectangular waveguide and coaxial probes have shown great potential for this purpose. However, to evaluate one parameter requires that the other be known a priori. This paper...
Conference Paper
Full-text available
The use of a combined embedded modulated scattering technique, utilizing a PIN diode-loaded dipole probe, and near-field microwave nondestructive testing technique, utilizing an open-ended rectangular waveguide, has been investigated as a means for determining the complex dielectric constant in which the MST probe is embedded. For this, the formula...
Article
Several approaches to using combined near-field microwave NDT techniques, utilizing open-ended rectangular waveguides, and embedded modulated scatterer technique (MST) for the determination of the dielectric properties of a material have been investigated in the past. A technique currently under investigation involves using the ratio of dynamic for...
Article
Full-text available
Detection of corrosion precursor pitting on metallic surfaces under various coatings and on bare metal is of keen interest in evaluation of aircraft fuselage. Near‐field microwave nondestructive testing methods, utilizing open‐ended rectangular waveguides and coaxial probes, have been used extensively for detection of surface flaws in metals, both...
Article
Full-text available
Fiber reinforced polymer (FRP) composites are increasingly being used for the rehabilitation of concrete structures. Detection and characterization of delaminations between an FRP composite and a concrete surface are of paramount importance. Consequently, the development of a one sided, non-contact, real time and rapid nondestructive testing (NDT)...
Article
The presence of a sufficient concentration of chloride ions at the steel/concrete interface can lead to the initiation of corrosion in reinforced concrete structures. To this end, a set of near-field microwave nondestructive testing experiments were conducted using three mortar specimens made with three different cements: ASTM Type I/II, III and V....
Article
Intrusion of chloride ions in steel reinforced concrete may lead to corrosion of steel reinforcement. This paper discusses a near-field microwave approach for detecting chloride intrusion in mortar exposed to cyclical chloride solution. Mortar specimens were exposed to tap water, as a control, and to a chloride solution of 2.8% salinity. Daily micr...
Article
Detection of corrosion, under paint and primer, in various metallic structural components, particularly when used in moist and salty environments is an important practical concern. Moreover, nondestructive testing techniques that do not require paint removal are desired. Near-field microwave nondestructive inspection techniques, employing open-ende...
Article
Microwave nondestructive techniques have been used in the past to detect the presence of corrosion in steel plates under layers of paint and primer. This paper describes a similar investigation for the detection of corrosion in aluminum plates. Dielectric properties of paint, primer and aluminum oxide were determined. Theoretical calculations produ...
Conference Paper
Full-text available
Investigation into the derivation of the dielectric property of a material with an embedded modulated PIN diode-loaded dipole is described, using an open-ended rectangular waveguide as the irradiating source. Previous measurements of the curing of a mortar specimen show the sensitivity of this combined technique and its potential for inspecting com...

Network

Cited By

Projects

Project (1)
Project
This project explores statistical, connectionist, and evolutionary techniques in computer music and art. It investigates fractal geometry (e.g., Zipf's law and power laws) in music information retrieval, computational aesthetics, and artificial creativity.