Scholarly Works, Mechanical Engineering

Permanent URI for this collection

https://hdl.handle.net/10919/23262

Research articles, presentations, and other scholarship

Browse

Now showing 1 - 15 of 15

Chain conformations and phase separation in polymer solutions with varying solvent quality
Huang, Yisheng; Cheng, Shengfeng (Wiley, 2021-10-02)
Molecular dynamics simulations are used to investigate the conformations of a single polymer chain, represented by the Kremer-Grest bead-spring model, in a solution with a Lennard-Jones liquid as the solvent when the interaction strength between the polymer and solvent is varied. Results show that when the polymer-solvent interaction is unfavorable, the chain collapses as one would expect in a poor solvent. For more attractive polymer-solvent interactions, the solvent quality improves and the chain is increasingly solvated and exhibits ideal and then swollen conformations. However, as the polymer-solvent interaction strength is increased further to be more than about twice the strength of the polymer-polymer and solvent-solvent interactions, the chain exhibits an unexpected collapsing behavior. Correspondingly, for strong polymer-solvent attractions, phase separation is observed in the solutions of multiple chains. These results indicate that the solvent becomes effectively poor again at very attractive polymer-solvent interactions. Nonetheless, the mechanism of chain collapsing and phase separation in this limit differs from the case with a poor solvent rendered by unfavorable polymer-solvent interactions. In the latter, the solvent is excluded from the domain of the collapsed chains while in the former, the solvent is still present in the pervaded volume of a collapsed chain or in the polymer-rich domain that phase separates from the pure solvent. In the limit of strong polymer-solvent attractions, the solvent behaves as a glue to stick monomers together, causing a single chain to collapse and multiple chains to aggregate and phase separate.
Extended Viterbi Algorithm for Hidden Markov Process: A Transient/Steady Probabilities Approach
Soltan, Reza A.; Ahmadian, Mehdi (Hikari Ltd., 2012)
In this paper an extended Viterbi algorithm is presented for first-order hidden Markov processes, with the help of a dummy combined state sequence. For this, the Markov switching’s transient probabilities and steady probabilities are studied separately. The algorithm gives a maximum likelihood estimate for the state sequence of a hidden Markov process. Comparing with the standard Viterbi algorithm, this method gives a higher maximum likelihood, and also picks up the state switching earlier, which is particularly important for the out of sample applications. The theory of this method is discussed in this paper and then a sample of a series of experiment is presented to illustrate the theory. A quantitative comparison is also given between this method and the standard Viterbi algorithm.
Guest editorial: Special Issue on Artificial Intelligence and Emerging Computational Approaches for Tribology
Zhang, Zhinan; Pan, Shuaihang; Raeymaekers, Bart (2024-04-02)
Here’s What I’ve Learned: Asking Questions that Reveal Reward Learning
Habibian, Soheil; Jonnavittula, Ananth; Losey, Dylan P. (Virginia Tech, 2021-07-02)
Robots can learn from humans by asking questions. In these questions the robot demonstrates a few different behaviors and asks the human for their favorite. But how should robots choose which questions to ask? Today’s robots optimize for informative questions that actively probe the human’s preferences as efficiently as possible. But while informative questions make sense from the robot’s perspective, human onlookers often find them arbitrary and misleading. For example, consider an assistive robot learning to put away the dishes. Based on your answers to previous questions this robot knows where it should stack each dish; however, the robot is unsure about right height to carry these dishes. A robot optimizing only for informative questions focuses purely on this height: it shows trajectories that carry the plates near or far from the table, regardless of whether or not they stack the dishes correctly. As a result, when we see this question, we mistakenly think that the robot is still confused about where to stack the dishes! In this paper we formalize active preference-based learning from the human’s perspective. We hypothesize that — from the human’s point-of-view — the robot’s questions reveal what the robot has and has not learned. Our insight enables robots to use questions to make their learning process transparent to the human operator.We develop and test a model that robots can leverage to relate the questions they ask to the information these questions reveal. We then introduce a trade-off between informative and revealing questions that considers both human and robot perspectives: a robot that optimizes for this trade-off actively gathers information from the human while simultaneously keeping the human up to date with what it has learned. We evaluate our approach across simulations, online surveys, and in-person user studies. We find that robots which consider the human’s point of view learn just as quickly as state-of-the-art baselines while also communicating what they have learned to the human operator. Videos of our user studies and results are available here: https://youtu.be/tC6y_jHN7Vw.
I Know What You Meant: Learning Human Objectives by (Under)estimating Their Choice Set
Jonnavittula, Ananth; Losey, Dylan P. (Virginia Tech, 2021-04-05)
Assistive robots have the potential to help people perform everyday tasks. However, these robots first need to learn what it is their user wants them to do. Teaching assistive robots is hard for inexperienced users, elderly users, and users living with physical disabilities, since often these individuals are unable to show the robot their desired behavior. We know that inclusive learners should give human teachers credit for what they cannot demonstrate. But today’s robots do the opposite: they assume every user is capable of providing any demonstration. As a result, these robots learn to mimic the demonstrated behavior, even when that behavior is not what the human really meant! Here we propose a different approach to reward learning: robots that reason about the user’s demonstrations in the context of similar or simpler alternatives. Unlike prior works — which err towards overestimating the human’s capabilities — here we err towards underestimating what the human can input (i.e., their choice set). Our theoretical analysis proves that underestimating the human’s choice set is risk-averse, with better worst-case performance than overestimating. We formalize three properties to generate similar and simpler alternatives. Across simulations and a user study, our resulting algorithm better extrapolates the human’s objective. See the user study here: https://youtu.be/RgbH2YULVRo.
Impacts of process-induced porosity on material properties of copper made by binder jetting additive manufacturing
Kumar, Ashwath Yegyan; Wang, Jue; Bai, Yun; Huxtable, Scott T.; Williams, Christopher B. (Elsevier, 2019-07-03)
Binder Jetting (BJ) is an efficient, economical, and scalable Additive Manufacturing (AM) technology that can be used in fabricating parts made of reflective and conductivematerials like copper, which have applications in advanced thermal and electrical components. The primary challenge of BJ is in producing fully dense, homogeneous partswithout infiltration. To this end, copper parts of porosities ranging from2.7% to 16.4%were fabricated via BJ, by varying powder morphology, post-process sintering, and Hot Isostatic Pressing conditions. The aim of this study is to characterize and quantify the effects of porosity on the material properties of Binder Jet pure copper parts. Copper parts with the lowest porosity of 2.7% demonstrated a tensile strength of 176 MPa (80.2% of wrought strength), a thermal conductivity of 327.9 W/m·K (84.5% that ofwrought copper), and an electrical conductivity of 5.6 × 107 S/m (96.6% IACS). The porosity-property relationship in these parts was compared against theoretical and empiricalmodels in the literature for similar structures. These studies contribute towards developing a scientific understanding of the process-property-performance relationship in BJ of copper and other printed metals, which can help in tailoring materials and processing conditions to achieve desired properties.
Learning to Share Autonomy Across Repeated Interaction
Jonnavittula, Ananth; Losey, Dylan P. (Virginia Tech, 2021-07-20)
Wheelchair-mounted robotic arms (and other assistive robots) should help their users perform everyday tasks. One way robots can provide this assistance is shared autonomy. Within shared autonomy, both the human and robot maintain control over the robot’s motion: as the robot becomes confident it understands what the human wants, it increasingly intervenes to automate the task. But how does the robot know what tasks the human may want to perform in the first place? Today’s shared autonomy approaches often rely on prior knowledge: for example, the robot must know the set of possible human goals a priori. In the long-term, however, this prior knowledge will inevitably break down — sooner or later the human will reach for a goal that the robot did not expect. In this paper we propose a learning approach to shared autonomy that takes advantage of repeated interactions. Learning to assist humans would be impossible if they performed completely different tasks at every interaction: but our insight is that users living with physical disabilities repeat important tasks on a daily basis (e.g., opening the fridge, making coffee, and having dinner). We introduce an algorithm that exploits these repeated interactions to recognize the human’s task, replicate similar demonstrations, and return control when unsure. As the human repeatedly works with this robot, our approach continually learns to assist tasks that were never specified beforehand: these tasks include both discrete goals (e.g., reaching a cup) and continuous skills (e.g., opening a drawer). Across simulations and an in-person user study, we demonstrate that robots leveraging our approach match existing shared autonomy methods for known goals, and outperform imitation learning baselines on new tasks. See videos here: https://youtu.be/NazeLVbQ2og.
Left-right tympanal size asymmetry in the parasitoid fly Ormia ochracea
Mikel-Stites, Max R.; Marek, Paul E.; Hellier, Madeleine E.; Staples, Anne E. (2024-08-02)
Ormia ochracea is a parasitoid fly notable for its impressive hearing abilities relative to its small size. Here, we use it as a model organism to investigate if minor size differences in paired sensory organs may be beneficial or neutral to an organism's perception abilities. We took high-resolution images of tympanal organs from 21 O. ochracea specimens and found a statistically significant surface area asymmetry (up to 6.88%) between the left and right membranes. Numerical experiments indicated that peak values of key sound localization variables increased with increasing tympanal asymmetry, which may explain features of the limited available physiological data.
MEMTRACK: A Deep Learning-Based Approach to Microrobot Tracking in Dense and Low-Contrast Environments
Sawhney, Medha; Karmarkar, Bhas; Leaman, Eric J.; Daw, Arka; Karpatne, Anuj; Behkam, Bahareh (2023)
Tracking microrobots is challenging, considering their minute size and high speed. As the field progresses towards developing microrobots for biomedical applications and conducting mechanistic studies in physiologically relevant media (e.g., collagen), this challenge is exacerbated by the dense surrounding environments with feature size and shape comparable to microrobots. Herein, we report Motion Enhanced Multi-level Tracker (MEMTrack), a robust pipeline for detecting and tracking microrobots using synthetic motion features, deep learning-based object detection, and a modified Simple Online and Real-time Tracking (SORT) algorithm with interpolation for tracking. Our object detection approach combines different models based on the object’s motion pattern. We trained and validated our model using bacterial micro-motors in collagen (tissue phantom) and tested it in collagen and aqueous media. We demonstrate that MEMTrack accurately tracks even the most challenging bacteria missed by skilled human annotators, achieving precision and recall of 77% and 48% in collagen and 94% and 35% in liquid media, respectively. Moreover, we show that MEMTrack can quantify average bacteria speed with no statistically significant difference from the laboriously-produced manual tracking data. MEMTrack represents a significant contribution to microrobot localization and tracking, and opens the potential for vision-based deep learning approaches to microrobot control in dense and low-contrast settings. All source code for training and testing MEMTrack and reproducing the results of the paper have been made publicly available https://github.com/sawhney-medha/MEMTrack.
Propagating fronts in fluids with solutal feedback
Mukherjee, Saikat; Paul, Mark R. (American Physical Society, 2020-03-25)
We numerically study the propagation of reacting fronts in a shallow and horizontal layer of fluid with solutal feedback and in the presence of a thermally driven flow field of counterrotating convection rolls. We solve the Boussinesq equations along with a reaction-convection-diffusion equation for the concentration field where the products of the nonlinear autocatalytic reaction are less dense than the reactants. For small values of the solutal Rayleigh number the characteristic fluid velocity scales linearly, and the front velocity and mixing length scale quadratically, with increasing solutal Rayleigh number. For small solutal Rayleigh numbers the front geometry is described by a curve that is nearly antisymmetric about the horizontal midplane. For large values of the solutal Rayleigh number the characteristic fluid velocity, the front velocity, and the mixing length exhibit square-root scaling and the front shape collapses onto an asymmetric self-similar curve. In the presence of counterrotating convection rolls, the mixing length decreases while the front velocity increases. The complexity of the front geometry increases when both the solutal and convective contributions are significant and the dynamics can exhibit chemical oscillations in time for certain parameter values. Last, we discuss the spatiotemporal features of the complex fronts that form over a range of solutal and thermal driving.
Radiation Search Operations using Scene Understanding with Autonomous UAV and UGV
Christie, Gordon A.; Shoemaker, Adam; Kochersberger, Kevin B.; Tokekar, Pratap; McLean, Lance; Leonessa, Alexander (Virginia Tech, 2016-08-31)
Autonomously searching for hazardous radiation sources requires the ability of the aerial and ground systems to understand the scene they are scouting. In this paper, we present systems, algorithms, and experiments to perform radiation search using unmanned aerial vehicles (UAV) and unmanned ground vehicles (UGV) by employing semantic scene segmentation. The aerial data is used to identify radiological points of interest, generate an orthophoto along with a digital elevation model (DEM) of the scene, and perform semantic segmentation to assign a category (e.g. road, grass) to each pixel in the orthophoto. We perform semantic segmentation by training a model on a dataset of images we collected and annotated, using the model to perform inference on images of the test area unseen to the model, and then re fining the results with the DEM to better reason about category predictions at each pixel. We then use all of these outputs to plan a path for a UGV carrying a LiDAR to map the environment and avoid obstacles not present during the flight, and a radiation detector to collect more precise radiation measurements from the ground. Results of the analysis for each scenario tested favorably. We also note that our approach is general and has the potential to work for a variety of diff erent sensing tasks.
Response to "Comments on "Design of active structural acoustic control systems by eigenproperty assignment [J. Acoust. Soc. Am. 99, 1785-1788 (1996)]
Burdisso, Ricardo A.; Fuller, Chris R. (Acoustical Society of America, 1996-03-01)
The authors thank Professor Cunefare for his interest in our paper on the design of active structural acoustic control systems. This Letter is to clarify questions raised in his Comments. (C) 1996 Acoustical Society of America.
A Review of Biosensors and Their Applications
Katey, Bright; Voiculescu, Ioana; Penkova, Anita Nikolova; Untaroiu, Alexandrina (ASME, 2023-11-06)
This paper reviews sensors with nano- and microscale dimensions used for diverse biological applications. A biosensor converts biological responses into electrical signals. In recent years, there have been significant advancements in the design and development of biosensors that generated a large spectrum of biosensor applications including healthcare, disease diagnosis, drug delivery, environmental monitoring, and water and food quality monitoring. There has been significant work to enhance the performance of biosensors by improving sensitivity, reproducibility, and sensor response time. However, a key challenge of these technologies is their ability to efficiently capture and transform biological signals into electric, optic, gravimetric, electrochemical, or acoustic signals. This review summarizes the working principle of a variety of biosensors in terms of their classification, design considerations, and diverse applications. Other lines of research highlighted in this paper are focused on the miniaturization of biosensing devices with micro and nano-fabrication technologies, and the use of nanomaterials in biosensing. Recently wearable sensors have had important applications such as monitoring patients with chronic conditions in home and community settings. This review paper mentions applications of wearable technology. Machine learning is shown to help discover new knowledge in the field of medical applications. We also review artificial intelligence (AI) and machine learning (ML)-based applications.
Supernova Physics at DUNE
Ankowski, Artur M.; Beacom, John; Benhar, Omar; Chen, Sun; Cherry, J. J.; Cui, Yanou; Friedland, Alexander; Gil-Botella, Ines; Haghighat, Alireza; Horiuchi, Shunsaku; Huber, Patrick; Kneller, James; Laha, Ranjan; Li, Shirley; Link, Jonathan M.; Lovato, Alessandro; Macias, Oscar; Mariani, Camillo; Mezzacappa, Anthony; O'Connor, Evan; O'Sullivan, Erin; Rubbia, Andre; Scholberg, Kate; Takeuchi, Tatsu (2016)
The DUNE/LBNF program aims to address key questions in neutrino physics and astroparticle physics. Realizing DUNE’s potential to reconstruct low-energy particles in the 10–100 MeV energy range will bring significant benefits for all DUNE’s science goals. In neutrino physics, low-energy sensitivity will improve neutrino energy reconstruction in the GeV range relevant for the kinematics of DUNE’s long-baseline oscillation program. In astroparticle physics, low-energy capabilities will make DUNE’s far detectors the world’s best apparatus for studying the electron-neutrino flux from a supernova. This will open a new window to unrivaled studies of the dynamics and neutronization of a star’s central core in real time, the potential discovery of the neutrino mass hierarchy, provide new sensitivity to physics beyond the Standard Model, and evidence of neutrino quantum-coherence effects. The same capabilities will also provide new sensitivity to ‘boosted dark matter’ models that are not observable in traditional direct dark matter detectors.
Terrain classification using intelligent tire
Khaleghian, Seyedmeysam; Taheri, Saied (Pergamon, 2017)
A wheeled ground robot was designed and built for better understanding of the challenges involved in utilization of accelerometer-based intelligent tires for mobility improvements. Since robot traction forces depend on the surface type and the friction associated with the tire-road interaction, the measured acceleration signals were used for terrain classification and surface characterization. To accomplish this, the robot was instrumented with appropriate sensors (a tri-axial accelerometer attached to the tire innerliner, a single axis accelerometer attached to the robot chassis and wheel speed sensors) and a data acquisition system. Wheel slip was measured accurately using encoders attached to driven and non-driven wheels. A fuzzy logic algorithm was developed and used for terrain classification. This algorithm uses the power of the acceleration signal and wheel slip ratio as inputs and classifies all different surfaces into four main categories; asphalt, concrete, grass, and sand. The performance of the algorithm was evaluated using experimental data and good agreements were observed between the surface types and estimated ones.

Browse

Browsing Scholarly Works, Mechanical Engineering by Content Type "Article"

Results Per Page

Sort Options