March 7, 2017

**Edgar Lobaton,** Assistant Professor, ECE Department, NCSU

"Geometric and Topological Mapping with Cyborg-Insect Networks"

Mapping of unknown environments is an essential and challenging task in search and rescue for emergency response application. Teams of small autonomous agents (e.g. biologically inspired milli-robots or cyborg insects known as *biobots*) have certain advantages over traditional platforms when performing exploration and mapping tasks since larger robotic systems may not be able to reach safely to locations under the rubble of collapsed buildings. Furthermore, biobotic platforms can be particularly versatile at locomotion and navigation in unstructured and dynamic scenes due to their natural ability to crawl through small spaces. Recent developments in neural engineering and neuromuscular stimulation have enabled us to directly control insect locomotion using wireless neuro-stimulators to enable remotely controlled biobots. Each agent is equipped with system-on-chip based ZigBee enabled wireless neuro-stimulation backpack system, remote navigational control circuits, inertial sensors and possibly environmental sensors (e.g., microphones or chemical).

In this talk, we present an approach for global exploration and mapping of unknown environments under minimal sensing and localization constraints. We exploit natural stochastic motion models and controlled locomotion of biobots in conjunction with an aerial leader to explore and map a domain of interest. Robust topological features from these local representations are extracted using topological data analysis and a simple classification scheme, and show asymptotic convergence of our approach. These maps are then merged into a global map, which can be visualized using a graphical representation that integrates geometric as well as topological features of the environment. Simulation and experimental results with biologically inspired robotic platform demonstrate the correctness of our approach. This mapping approach relies on the robust identification of certain modes of locomotion of the biobots and the local identification of proximity between the agents. We illustrate how data streams from the on-board inertial sensors can identify different modes of locomotion / activity, and hence enhancing localization and mapping.

November 1, 2016

Kristin S. Miller, Biomedical Engineering, Tulane University

"Leveraging Growth and Remodeling Models to Improve Women's Health"

Briefly, our work seeks to leverage experimental and computational biomechanics to determine the role of each load-bearing constituent within the female reproductive tract. In particular, we are interested in the role of elastic fibers, which we hypothesize significantly contribute to clinical issues in women's health, such as pelvic organ prolapse and preterm birth. Our long term goal is to leverage predictive mathematical models to delineate the mechanical etiology of prolapse and preterm birth, as well as to design effective clinical interventions to reduce their incidence.

October 18, 2016

John Lagrone, CCS postdoc, Tulane University

"Applications of Complete Radiation Boundary Conditions to Electromagnetic and Elastic Problems"

An important issue in the simulation of waves is the ability to truncate unbounded domains into regions of interest that can be simulated efficiently and accurately for long times. In the context of finite difference time domain electromagnetics we demonstrate that complete radiation boundary conditions (CRBC) can be implemented as a Double Absorbing Boundary. This results in a thin, non-reflecting layer which inherits the desirable properties of CRBCs, namely a clear notion of convergence and an a priori error estimate that allows for the selection of optimized parameters. The performance of the method is demonstrated with numerical experiments. We also briefly discuss ongoing work concerning the application of CRBCs to (an)isotropic elastic waves.

September 27, 2016

Professor Shilpa Khatri, UC Merced

Fluid-structure interactions: Applications to marine phenomena

To understand the fluid dynamics of marine phenomena, for example particles settling, droplets rising, and pulsating coral, fluid-structure interaction problems must be solved. Challenges exist in developing analytical and numerical techniques to solve these complex flow problems with boundary conditions at fluid-structure (solid and porous) interfaces. I will present details of two different problems where these challenges are handled: (1) modeling of marine aggregates settling in density stratified fluids and (2) modeling of pulsating soft corals. These problems will be motivated by field and experimental work in the marine sciences. I will discuss these related data and provide comparisons with the modeling.

April 26, 2016

Professor Martin Frank, RWTH Aachen University, Germany

"Asymptotic preserving schemes for radiation transport, with applications to radiotherapy dose calculation"

The biggest challenge for radiation transport simulations are multi-scale media, with strongly varying material coefficients. Especially important is that a scheme captures the singular limit of strong scattering. This is relevant in many applications, among them radiotherapy dose calculation. Schemes that behave well in this limit are called asymptotic-preserving. We present a simple numerical scheme to solve the PN equations. Because of its simplicity, the scheme can be implemented very efficiently. We present the idea of the scheme, sketch a stability proof, and show that it is asymptotic-preserving. Several numerical tests are shown, especially applications to radiotherapy dose calculation. The code can be downloaded so that all results can be reproduced.

January 26, 2016

Alvaro Vazquez-Mayagoitia, Argonne Leadership Computing Facility (ALCF)

"Many-core Technologies to Accelerate Quantum Chemistry Calculations"

The availability of modern massive parallel computers has allowed scientists to study larger and more complex problems. Nowadays, simulations of unprecedented size can be performed, and new models with more realistic conditions can be tested and applied faster than just a few years ago. Computer architectures are constantly evolving and increasing calculation power, consequently posing challenges to computer and computational scientists.

One of those challenges is that efficient use of new advanced computers require sophisticated algorithms and the application of modern programming paradigms that allow researchers to speed up new discoveries and inventions. Leadership computing facilities serve as a link among scientists, applications and computers, with the aim to facilitate the transition to impeding technologies and to overcome new computational challenges. In this talk, I will discuss the role of Argonne Leadership Computing Facility in supporting scientific projects with parallel computing and novel computer science tools. In particular, to accelerate the exploration polymorphs of molecular crystals using atomistic models and quantum chemistry.

Bio: Alvaro Vazquez-Mayagoitia is currently an Assistant Computational Scientist at the Argonne Leadership Computing Facility. He is an experienced developer of several electronic structure codes on high performance computers. Lately, Alvaro has been focused on the enhancement of capabilities in quantum chemistry codes, such as MADNESS and NWChem, of which he is a co-author. His experience embraces density functional theory (DFT) and quantum chemistry methods.

January 12, 2016

Eva Kanso, Associate Professor, Aerospace and Mechanical Engineering, University of Southern California

"Shock Waves in Confined Microswimmers"

Active systems, i.e., systems driven internally by self-propelled individual units, often exhibit rich collective behavior at the system’s scale; a scale that is typically several orders of magnitude larger than the scale of the individual unit. Such collective behavior naturally arises in disparate biological systems, from schools of fish to populations of bacterial cells. The emergence of highly coordinated collective motions in bacterial cells and self-propelled particles in viscous fluids is a well studied phenomenon, mostly in relation to the instabilities and spatiotemporal fluctuations in three-dimensional (3D) systems. Recently, attention began to shift to the collective dynamics of particles confined in quasi two-dimensional (2D) geometries, especially in light of the rapid technological advances in microfluidics. Geometric confinement changes drastically the nature of the hydrodynamic interactions among microswimmers. In 3D, the long-ranged hydrodynamic interactions are driven by force dipoles exerted by self-propelled particles on the fluid medium. In quasi-2D geometries, the confining walls screen the force dipole contribution, making it subdominant in comparison with the potential dipole arising from incompressibility. As a result, the long-range interactions among swimmers can be obtained from the superposition of dipole singularities.

In this talk, we present a hydrodynamic dipole model for confined microswimmers and show surprising results in their collective dynamics. For example, we show hydrodynamically-triggered transitions in confined microswimmers from turbulent-like swimming to aggregation and clustering. The collective dynamics is even more interesting in narrow channels, where we observe the emergence of compression and expansion density shock waves. This behavior is the result of a non-trivial interplay between hydrodynamic interactions and geometric confinement, and is confirmed by a novel quasilinear wave model that properly captures the dependence of the shock formation on the external flow. These findings can be applied to guide the development of novel mechanisms for controlling the emergent behavior of particles in microfluidic channels, thus enabling processes such as sorting of cells in flow channels.

October 6, 2015

Christina Hamlet, Mathematics and CCS, Tulane

"Making Waves: Neural activation and sensory feedback drive an integrative model of a swimming lamprey"

Abstract: The lamprey is a jawless fish that serves as a model for both neurophysiology and locomotion studies. While their anguilliform (eel-like) swimming involves the seemingly simple action of sending traveling waves of curvature from head to tail, this motion requires the coordination of several systems across multiple scales from electrical signals to the fluid environment. I will present a 2D, integrative, multi-scale model of the lamprey's swimming driven by neural activation from a central pattern generator (CPG) modeled as a chain of coupled oscillators. The CPG in turn drives muscle kinematics and fully coupled fluid-structure interactions implemented in an immersed boundary framework to produce the emergent swimming mode. Body curvature and rate of curvature change provide feedback to the CPG. Effects on the speed, stability and cost (metabolic work) of swimming due to the nonlinear dependencies associated with muscle force development combined with feedback to the activation wave are estimated and examined.

September 22, 2015

Amy Buchmann, Mathematics and CCS, Tulane

"Mathematical and Computational Modeling of Bacterial Motility and Swarming"

Computational models play an important role in understanding bacterial movement. For example, the very social *Myxococcus xanthus*, a bacterium commonly found in soil and known for its multicellular interactions, can be modeled using the subcellular element method. I will present an implementation of this model and show how it can be used to study the effects of cell flexibility, cell-cell adhesion, and cellular reversal periods on cell-cell interactions. To characterize cell-cell interactions, the contacts between cells in simulations are analyzed to determine how these properties influence the populations' ability to form and keep cell-cell connections. Bacterial flagella may also play an important role in the development of microfluidics devices. Recent experimental work has suggested that the flagella of bacteria may be used as motors in microfluidics devices by creating a bacterial carpet. I will show how the flow induced by bacterial carpets can be modeled and analyzed using the method of regularized Stokeslets, and also examine the transport of vesicles of finite size by arrays of rotating flagella.

September 8, 2015

Alex Hoover, Mathematics and CCS, Tulane

"From Pacemaker to Vortex Ring: Modeling Jellyfish Propulsion and Maneuvering"

With a fossil record dating over half a billion years, jellyfish represent one of the earliest examples of how multicellular organisms first organized into moving systems. Lacking an agonist-antagonist muscle pairing, jellyfish swim via a process of elastic deformation and recoil. Jellyfish propulsion is generated via the coordinated contraction of its elastic bell by its coronal swimming muscles and a complementary re-expansion that is passively driven by stored elastic energy. Recent studies have found jellyfish to be one of the most efficient swimmers due to its low energy expenditure in their forward movement. Using an immersed boundary framework, we will further examine the performance of jellyfish swimming by incorporating material models that are informed by the musculature present in jellyfish into a model of the elastic jellyfish bell in three dimensions. The fully-coupled fluid structure interaction problem is solved using an adaptive and parallelized version of the immersed boundary method (IBAMR). This model is then used to explore how mechanical and scaling properties of the bell affect the work done by the bell as well as the cost of transport related to jellyfish locomotion. We then shift our focus onto how the underlying acephalic neuromuscular organization of their bell allows for complicated swimming behaviors, such as steering and maneuvering.

April 21, 2015

Joanna Gyory, EEB and CCS, Tulane

"Simulations of blue crab larval dispersal in the northern Gulf of Mexico"

The commercial blue crab fishery is the 9th most valuable in the United States. In 2013, fishers harvested more than 60,000 metric tons of crabs that sold for $191 million. However, blue crab abundance varies greatly from year to year, and the causes are not fully understood. One important driver of this variability may be the dispersal dynamics of the larval stage. Blue crab larvae spend ~30-60 days at the ocean surface, where they are transported by currents until they metamorphose into juvenile crabs. We coupled an ocean circulation model with a particle-tracking model to simulate larval dispersal in the northern Gulf of Mexico, and then used network metrics to determine how blue crab populations from different estuaries may be interconnected. We found that the Mississippi River Delta is a barrier to larval dispersal, and that a high percentage of larvae were retained near the parental estuary. Population connectivity declines throughout the spawning season, probably due to changes in near-shore ocean circulation patterns. We also investigated the potential impacts of the Deepwater Horizon oil spill, and found that more than 96% of larvae east of the Mississippi River Delta may have been exposed to oil. We are now using the models to examine how the annual low-oxygen Dead Zone in the Gulf of Mexico might change larval dispersal patterns by restricting the spawning sites available to adult females.

October 14, 2014

Howard Elman, Computer Science and Institute for Advanced Computer Studies, University of Maryland

https://www.cs.umd.edu/users/elman/"Efficient Solution Algorithms for Stochastic Partial Differential Equations"

We consider new computational methods for solving partial differential equations (PDEs) when components of the problem such as diffusion coefficients or boundary conditions are not known with certainty but instead are represented as random fields. In recent years, several computational techniques have been developed for such models that offer potential for improved efficiencies compared with traditional Monte-Carlo methods. These include stochastic Galerkin methods, which use an augmented weak formulation of the PDE derived from averaging with respect to expected value, and stochastic collocation methods, which use a set of samples relatively small in cardinality that captures the character of the solution space. We give an overview of the relative advantages of these two methods and present efficient computational algorithms for solving the algebraic systems that arise from them. In addition, we show that these algorithms can be combined with techniques of reduced-order modeling to significantly enhance efficiency with essentially no loss of accuracy.

September 30, 2014

Nicole Gasparini, Earth and Environmental Sciences, Tulane University

"Landlab: An open-source landscape evolution model"

The Landlab project creates an environment in which scientists can build a numerical landscape model without having to code all of the individual components. Landscape models compute flows of mass, such as water, sediment, glacial ice, volcanic material, or landslide debris, across a gridded terrain surface. Landscape models have a number of commonalities, such as operating on a grid of points and routing material across the grid. Scientists who want to use a landscape model often build their own unique model from the ground up, re-coding the basic building blocks of their landscape model rather than taking advantage of codes that have already been written.

Whereas the end result may be novel software programs, many person-hours are lost rewriting existing code, and the resulting software is often idiosyncratic and not able to interact with programs written by other scientists in the community. This individuality in software programs leads to lost opportunity for exploring an even wider array of scientific questions than those which can be addressed using a single model. The Landlab project seeks to eliminate these redundancies and lost opportunities by creating a user- and developer-friendly numerical landscape modeling environment which provides scientists with the fundamental building blocks needed for modeling landscape processes. The model code is written in Python, an open-source language that is a relatively easy language for casual programmers to learn.

September 23, 2014

Noa Marom, Physics & Engineering Physics, Tulane

"Structure Matters More Than Size: Tuning the Electronic Properties of TiO_{2} Clusters"

Atomic clusters comprising up to few tens of atoms offer exciting prospects for designing nano-catalysts, owing to their high reactivity and the strong dependence of their electronic properties on their size and structure. Quantum mechanical atomistic simulations may be used for computer-aided design of cluster-based nano-catalysts.

The vast majority of computational studies of clusters have focused on finding their global minimum structure, using various global optimization techniques. However, the most stable structures of atomic clusters are not necessarily optimal for catalysis. Rather, the presence of an active site may enhance the reactivity and/or selectivity of a cluster-based nano-catalyst. The presence of such an active site may be associated with certain electronic properties. For example, clusters with a high electron affinity or a low ionization potential would be more likely to accept or donate an electron.

We use a basin hopping algorithm based on density functional theory (DFT) in combination with many-body perturbation theory to show that photoemission spectroscopy (PES) experiments on (TiO_{2})_{2-10} anions select for clusters with a high electron affinity, rather than the most stable isomers. We then develop a suite of property-based genetic algorithms (GA) tailored to optimize for low-energy (EGA), high vertical electron affinity (VEA-GA), and low vertical ionization potential (VIP-GA). Analysis of the best structures found by the VEA-GA and VIP-GA reveals structure-property relations and explains the absence of the expected size trends. A high VEA is associated with a large number of dangling O atoms. A low VIP is associated with a low connectivity between dangling O atoms. These structural features become less probable with increasing cluster size, explaining why some smaller clusters have higher VEAs, lower VIPs, and narrower fundamental gaps than larger clusters.

References:

[1] N. Marom, M. Kim, and J. R. Chelikowsky, Phys. Rev. Lett. 108, 106801 (2012)

[2] N. Marom, J. E. Moussa, X. Ren, A. Tkatchenko, and J. R. Chelikowsky, Phys. Rev. B 84, 245115 (2011)

[3] S. Bhattacharya, B. H. Sonin, C. J. Jumonville, L. M. Ghiringhelli, N. Marom, to be published

Center for Computational Science, Stanley Thomas Hall 402, New Orleans, LA 70118 ccs@tulane.edu